Full text of "The Texas journal of science"

See other formats

<T4X
A) f4

PUBLISHED QUARTERLY BY
THE TEXAS ACADEMY OF SCIENCE

THE

TEXAS JOURNAL

SCIENCE

smithso}^

Volume 59
Number 1
February 2007

GENERAL INFORMATION

MEMBERSHIP. -Any person or member of any group engaged in
scientific work or interested in the promotion of science is eligible for
membership in The Texas Academy of Science. For more informa¬
tion regarding membership, student awards, section chairs and vice¬
chairs, the annual March meeting and author instructions, please ac¬
cess the Academy's homepage at:

www.texasacademyofscience.org

Dues for regular members are $30.00 annually; supporting mem¬
bers, $60.00; sustaining members, $100.00; patron members, $150.00;
associate (student) members, $15.00; family members, $35.00; affili¬
ate members, $5.00; emeritus members, $10.00; corporate members,
$250.00 annually. Library subscription rate is $50.00 annually.

The Texas Journal of Science is a quarterly publication of The
Texas Academy of Science and is sent to most members and all sub¬
scribers. Payment of dues, changes of address and inquiries regarding
missing or back issues should be sent to:

Dr. Fred Stevens, Executive Secretary
The Texas Academy of Science
CMB 6252
Schreiner University
Kerrville, Texas 78028-5697
E-mail: FStevens@schreiner.edu

The Texas Journal of Science (ISSN 0040-4403) is published quarterly at Lawrence, Kansas
(Allen Press), U.S.A. Periodicals postage paid at San Angelo, Texas and additional mailing
offices. POSTMASTER: Send address changes and returned copies to The Texas Journal of
Science, Dr. Fred Stevens, CMB 6252, Schreiner University, Kerrville, Texas 78028-5697, U.S.A.
The known office of publication for The Texas Journal of Science is the Department of Biology,
Angelo State University, San Angelo, Texas 76909; Dr. Ned E. Strenth, Managing Editor.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval
system or transmitted, in any form or by any means, electronic, mechanical, recording or
otherwise, without the prior permission of the Managing Editor of the Texas Journal of Science.

THE TEXAS JOURNAL OF SCIENCE

Volume 59, No. 1

Feburary, 2007

CONTENTS

Status of the Feral Burro {Equus asinus) in Trans-Pecos Texas.

By Frederick B. Stangl, Jr., William B. Cook, Norman V. Horner

and Greg H. Broussard..... . . . . . . . . . 3

Habitat Attributes and Population Size of Texas Kangaroo Rats on an
Intensely Grazed Pasture in Wichita County, Texas.

By Jim R. Goetze, William C. Stasey, Allan D. Nelson

and Philip D. Sudman..... . . . . . . . . . . . 11

Infestation by Chigger Mites in Two Lizard Species from a Dune Habitat
of Northern Mexico.

By Cristina Garcia-De la Pena, Gamaliel Castaneda

and Cameron W. Barrows . . . . . 23

Helminth Parasite Assemblages in Bullfrogs {Rana catesbeiana)
from Southeast Texas.

By H. Randall Yoder and G. Whitney Gomez . . . . 33

Inventory of Three Beetle Species Assemblages (Coleoptera: Carabidae
Scarabaeoidea, Tenebrionoidea) from the Chihuahuan Desert of West Texas.

By Stephanie M. Middleton, Greg H. Broussard, Michael M. Shipley

and Roy C. Vogtsberger . . . . . . . . . . 39

Effect of Human Disturbance on the Abundance and Spatial Distribution of the
Atlantic Ghost Crab {Ocypode quadrata) (Fabricius 1798) on a Texas Beach.

By Alan D. Mac car one & Patrick L. Mathews . . . . . . . . . . . . . 51

GPS-Based Analysis of Shoreline Change, 1995-2005, Mad Island Marsh Preserve,
Matagorda County, Texas.

By Webster Mangham and Harry Williams . . . . . 61

General Notes

Demise of an Introduced Population of the Sailfm Molly, Poecilia latipinna.

in the Upper Guadalupe River of Central Texas.

By Fred B. Stevens, Adriana S. Jarussi and James Athey..... . . . 73

Membership Application . . . . . 76

Author Instructions

THE TEXAS JOURNAL OF SCIENCE
EDITORIAL STAFF

Managing Editor:

Ned E. Strenth, Angelo State University
Manuscript Editor:

Frederick B. Stangl, Jr., Midwestern State University
Associate Editor for Botany:

Janis K. Bush, The University of Texas at San Antonio
Associate Editor for Chemistry:

John R. Villarreal, The University of Texas-Pan American
Associate Editor for Computer Science:

Nelson Passos, Midwestern State University
Associate Editor for Environmental Science:

Thomas LaPoint, University of North Texas
Associate Editor for Geology:

Ernest L. Lundelius, University of Texas at Austin
Associate Editor for Mathematics and Statistics:

E. Donice McCune, Stephen F. Austin State University
Associate Editor for Physics:

Charles W. Myles, Texas Tech University

Manuscripts intended for publication in the Journal should be submitted in
TRIPLICATE to:

Dr. Frederick B. Stangl, Jr.

TJS Manuscript Editor
Department of Biology
Midwestern State University
Wichita Falls, Texas 76308
frederick.stangl@mwsu.edu

Scholarly papers reporting original research results in any field of
science, technology or science education will be considered for publication in
The Texas Journal of Science. Instructions to authors are published one or
more times each year in the Journal on a space-available basis, and also are
available on the Academy's homepage at:

www.texasacademyofscience.org

AFFILIATED ORGANIZATIONS
American Association for the Advancement of Science,

Texas Council of Elementary Science
Texas Section, American Association of Physics Teachers
Texas Section, Mathematical Association of America
Texas Section, National Association of Geology Teachers
Texas Society of Mammalogists

TEXAS J. SCI. 59(1):3-10

FEBRUARY, 2007

STATUS OF THE FERAL BURRO (EQUUS ASINUS)

IN TRANS-PECOS TEXAS

Frederick B. Stangl, Jr., William B. Cook, Norman V. Horner
and Greg H. Broussard*

Department of Biology, Midwestern State University
Wichita Falls. Texas 76308
"^Current address:

Department of Entomology, Oklahoma State University
Stillwater, Oklahoma 74078

Abstract.-The feral burro (Equus asinus) occurs across much of the rugged
landscapes of the western United States. Recent sightings in Brewster and Presidio
counties of Trans-Pecos Texas indicate that free-ranging populations are now locally
established in parts of the Big Bend region of Texas. A brief historical perspective of
E. asinus in the state is provided along with a discussion of possible ecological
implications of its presence. Given the potentially negative impact of E. asinus on
livestock and resident big game species, it is proposed that a detailed assessment of
distribution and population numbers in west Texas is presently warranted.

The feral burro {Equus asinus) is native to the “broken,
undulating, stony desert country” of North Africa and the Middle
East (Nowak 1999). Domesticated about 6,000 years ago, its
durability and endurance as a draft and pack animal led to its use on
every continent except Antarctica. Spanish explorers first intro¬
duced the burro to the Americas in the sixteenth century, where the
species soon became an integral and necessary commodity. The
development of mechanized transportation rendered the burro
obsolete in the United States by the earliest 1900s, after which time
released and escaped animals quickly became established in several
western states, mostly on the extensive tracts of BLM (Bureau of
Land Management) properties and other federal lands (Jenkins &
Ashley 2003).

Free-ranging populations readily adapted to parts of the desert
southwest (notably Arizona, Nevada, and southern California), and
have proliferated under the protection of the Wild and Free-
Roaming Horse and Burro Act of 1971. However, E. asinus seems
to have escaped the notice of regional faunal surveys in Texas over
the past century (Bailey 1905; Blair 1940; Borell & Bryant 1942;

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

Davis 1944; Taylor & Davis 1947; Davis 1978; Davis & Schmidly
1994; Stangl et al. 1994; Goetze 1998; Schmidly 1977, 2002,
2004).

Findley (1978) indicated that a single population occurs in
Lincoln County of central New Mexico, but Jenkins & Ashley
(2003) do not map Texas as within the known distribution of the
feral burro. In fact, the first literature record of the feral burro in
Texas of which the authors are aware is Yancey’s (1997) observa¬
tion of a solitary burro sighting from a helicopter over the north¬
western panhandle of Big Bend Ranch State Park in southern
Presidio County of Trans-Pecos Texas.

Yancey (1997) interpreted his observation as representing an
apparently rare feral population on the state park. His sighting
might just as well have been discounted as a domesticated vagrant
or escapee, for the species is widely bred and maintained in Texas
and elsewhere as a pack animal for hunters and campers, as a
coyote deterrent among sheep and goats, or simply as a novelty
item.

Burros and burro sign (Figure 1) were first observed in 1999 on
the Midwestern State University’s 518 ha Dalquest Research Site,
about 50 km southeast of Yancey’s (1997) observation, and have
since been regularly observed by bimonthly university field parties.
This property, donated to the university in 1996 by the late Walter
W. Dalquest and wife Rose, straddles the Brewster/Presidio county
line, and borders the Big Bend Ranch State Park to the south.
Terrain consists of heavily eroded canyonlands spreading from the
eastern edge of Bandera Mesa, and is bisected by an intermittent
stream - the Alamo de Cesario. Two semipermanent springs
(Lower Spring - 29" 33.391 N, 103" 47.133 W; elevation 1149 m;
and Upper Spring - 29" 33.165 N, 103" 47.151 W; elevation 1150
m) feed into the stream. Flood plains of this watercourse and
associated tributaries support stretches of riparian woodlands,
including mesquite (Prosopis glandulosa), gray oak {Quercus

STANGL ET AL.

Fig. 1. Photograph of four burros taken on 7 July 2001 on south fork of Alamo de
Cesario Creek, on Midwestern State University’s Dalquest Research Site, Presidio
Co., Texas. (Photo by G. H. Broussard)

grisea), little walnut {Juglans microcarpa), graythom lotebush
(Ziziphus obtusifolia), catclaw mimosa {Mimosa biuncifera),
allthom {Koeberlinia spinosa), and desert sumac {Rhus
microphylla).

At least one herd of 8-10 individuals frequents the Dalquest
Research Site and adjoining private ranchlands. Sightings, tracks
(including the silver dollar-sized tracks of newborn foals noted in
June 2006) and dung indicate concentrated activity around the
springs. These observations and that of Yancey (1997) clearly
indicate to the authors that the feral burro is now a viable
component of the Texas mammalian fauna. Its reputation as a
potentially destructive ecological agent in more westerly deserts
(e.g., Hanley & Brady 1977; Jones 1985; Nowak 1999; Jenkins &
Ashley 2003) prompted an inquiry of university biologists and state
wildlife professionals with extensive Trans-Pecos field experience.

The occurrence of the feral burro is presently localized, but
aspects of the species’ biology suggest that the species is worthy of
attention by wildlife biologists. This is a large (to 230 kg), social
animal that has no effective predators in the American deserts
(Hoffmeister 1986; Jenkins & Ashley 2003). Additionally, sur-

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

vivorship probabilities of about 90% for most age classes, low foal
mortality rates, and few disease problems (Ruffner & Carothers
1982; Jenkins & Ashley 2003) have permitted many feral
populations in other western states to double in number every 5-6
years (Ruffner & Carothers 1982).

The Feral Burro in Texas

The authors owe Louis Armendirez for the following account of
the history of the feral burro in Texas. Presently superintendent of
the Big Bend Ranch State Park, he was bom and raised in Brewster
and Presidio counties, and has recollections of burros dating to his
childhood days along the Rio Grande in the mid- 1900s. During the
1950s, burros were common as both free-ranging livestock and as
feral animals. Low river levels during periods of drought permitted
free interchange between herds on either side of the river.
However, a demand for dog food led to mass roundups and
slaughter of feral animals during the 1950s, which nearly
eliminated local populations.

The feral burrow remained uncommon until the past decade, at
which time numbers increased and the range in Texas expanded,
promoted largely by immigration from Mexican herds. This timing
coincides with the appearance of the species along the northern
reaches of Big Bend Ranch State Park and the Dalquest Research
Site. Presently, the Trans-Pecos distribution appears to be mostly
restricted along and within 40 km of the “big bend” of the Rio
Grande (Figure 2), with sightings from Candelaria of Presidio
County (M. Sullins, pers. comm.) eastward to Horse Canyon of
Black Gap Wildlife Management Area of Brewster County (M.
Haiduk, pers. comm.). Most sightings are of solitary animals or
small groups of a dozen or less, but at least one herd in excess of 50
animals presently ranges over southern parts of the Big Bend Ranch
State Park (L. Armendirez, pers. comm.), where the animals seem
most abundant. Feral animals also occur in lesser numbers on Big
Bend National Park and adjoining private ranches (B. Tarrant & M.
Sullins, pers. comm.).

STANGL ET AL.

Fig. 2. Map of Big Bend region of southern Trans-Pecos Texas, including state and
federal land holdings referenced in text: Big Bend National Park (BBNP), Big Bend
Ranch State Park (BBRSP), Black Gap Wildlife Management Area (BGWMA), and
MSU’s Dalquest Research Site (DRS).

Ecological Implications

The comparatively narrow inhabited zone of the feral burro in
Texas likely reflects the recent nature of the species’ movements
northward into the interior of the Trans-Pecos. It seems likely that
further incursions beyond the state and federal lands will depend on
the tolerance of private landowners, who can choose to control or
eliminate populations at their discretion.

Studies of the feral burro in the deserts of southern California
and Arizona (Woodward & Ohmart 1976; Hanley & Brady 1977;
Hoffmeister 1986) indicate that the species can have a substantial
negative impact on vegetation and terrain during seasonal shifts in
foraging behavior. Animals wander washes for spring greenery,
remain more localized near water sources in summer, and range
farther outwards to forage during the cooler fall and winter months.
The generally negative perception of the ecological impact of E.
asinus has prompted a series of position statements from at least
one conservation group (Sierra Club 2006) that include a call for
the elimination of the burro from national parks and monuments.

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

and from other public lands where it poses a threat to rare,
endangered, threatened, or endemic species of plants and animals.

Burros and livestock-Vxiov experience with west Texas land-
owners suggests a commendable attitude of tolerance for wildlife
species, both native and introduced. However, this attitude of
“benign neglect” could change in short order, if economic interests
are threatened.

Because the burro appears to be largely restricted to state and
federal lands at this time, the feral burro population is not an
immediate issue with local ranching interests. Habitat preferences
and foraging habits of E. asinus typically do not coincide with
optimum cattle grazing areas, suggesting that it is not a serious
competitor with cattle, at least when resources are abundant
(Jenkins & Ashley 2003). However, Trans-Pecos range conditions
are seldom better than marginal, and a dry year may not permit a
section of land to support more than a few head of cattle. At such
times, any outside competition with livestock would not be
welcomed.

Burros and bighorns game species are viewed as
increasingly valuable cash crops by Trans-Pecos ranchers, who
lease out hunting rights. Due to an extensive and long-term
cooperative effort between private landowners and Texas Parks &
Wildlife Department, the position of the mule deer {Odocoileus
hemionus) as the premier big game species in west Texas is being
challenged by reintroduced populations of the desert bighorn {Ovis
canadensis).

It is the possible competitive superiority of the burro with the
desert bighorn that probably will most concern Texas landowners
and wildlife biologists. Jones (1985) cautions that much of the
evidence of the burro’s negative impact on the bighorn appears to
be circumstantial and anecdotal, but there seems to be no question
that the feral burro is capable of excluding the sheep from the
limited and localized water and foraging sites, especially when
these resources are limited (Hoffmeister 1986; Nowak 1999;
Jenkins & Ashley 2003).

STANGL ET AL.

Any perceived competition with either livestock or bighorn
sheep by the burro for water and scant forage - both forbs and
grasses - would likely not be tolerated by ranchers. However,
control of the feral burro population on public lands is a more
complicated matter, given the public perception of the burro as a
romantic relic of the Old West, and the resulting protection afforded
by the Wild and Free-Roaming Horse and Burro Act of 1971.

Although the presence of the feral burro in Texas is presently a
localized phenomenon, there seems little to impede the species’
proliferation and expansion across the rugged Trans-Pecos terrain.
Given the reproductive potential of this large ungulate, it would
seem prudent at this time to initiate a coordinated effort by wildlife
agencies for the purpose of documenting numbers and distribution
of feral populations. Only with such data can any appropriate and
informed management strategy be initiated.

Acknowledgments

Matt Williams first brought the burros on the Dalquest Research
Site to the senior author’s attention. We gratefully acknowledge the
following colleagues who have shared their knowledge and
observations (or lack thereof) of feral burros in Texas: Big Bend
Ranch State Park superintendent Louis Armendirez; Texas Parks &
Wildlife biologists Billy Tarrant for Presidio County and Mike
Sullins for Brewster County, Mike Haiduk of Lamar University;
Loren Ammerman of Angelo State University, and David Schmidly
of Oklahoma State University. We thank Jon Baskin and an
anonymous reviewer for their contributions to an earlier draft of the
manuscript.

Literature Cited

Bailey, V. 1905. Biological Survey of Texas. North American Fauna, 25:1-222.

Blair, W. F. 1940. A contribution to the ecology and faunal relationships of the
mammals of the Davis Mountain region, southwestern Texas. Miscellaneous
Publications, Museum of Zoology, University of Michigan, 46:1-39.

Borell, A. E., & M. D. Bryant. 1942. Mammals of the Big Bend area of Texas.

University of California Publications in Zoology, 48:1-62.

Davis, W. B. 1944. The mammals of Culberson County, Texas. Journal of Mam¬
malogy, 25:254-273.

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

Davis, W. B. 1978. The Mammals of Texas. Texas Parks & Wildlife Department,
Austin, Bulletin 41:1-294.

Davis, W. B. & D. J. Schmidly. 1994. The Mammals of Texas. Texas Parks & Wildlife
Press, Austin, x + 338 pp.

Findley, J. S. 1987. The Natural History of New Mexican Mammals. University of New
Mexico Press, Albuquerque, xii + 164 pp.

Goetze, J. R. 1998. The mammals of the Edwards Plateau, Texas. Special Publications,
Museum of Texas Tech University, 41:1-263.

Hanley, T. A., & W. W. Brady. 1977. Feral Burro Impact on a Sonoran Desert Range.
Journal of Wildlife Management, 30:374-377.

Hoffmeister, D. F. 1986. Mammals of Arizona. University of Arizona Press, Tuscon,
xix + 602 pp.

Jenkins, S. H., & M. C. Ashley. 2003. Wild Horse: Equus caballus and allies. Pp. 1 148-
1163, in Wild Mammals of North America: Biology, Management, and
Conservation (G. A. Feldhammer, B. C. Thompson, & J. A. Chapman, eds.). Johns
Hopkins University Press, Baltimore, xiii + 1216 pp.

Jones, F. L. 1985. Competition. Pp. 197-216, in The Desert Bighorn: Its Life
History,Ecology, and Management (G. Monson & L. Sumner, eds.). University of
Arizona Press, Tuscon, xxi + 370 pp.

Nowak, R. M. 1999. Horses, Zebras, and Asses. Pp. 1008-1025, w Walker’s Mammals
of the World, 6* Edition. Johns Hopkins University Press, Baltimore, 2:837-1936.

Ruffner, S. A., & S. W. Carothers. 1982. Age structure, condition and reproduction of
two populations of Equus asinus (Equidae) from Grand Canyon National Park,
Arizona. Southwestern Naturalist, 27(4):403-41 1 .

Schmidly, D. J. 1977. The Mammals of Trans-Pecos Texas. Texas A&M University
Press, College Station, xiii + 225 pp.

Schmidly, D. J. 2002. Texas Natural History: A Century of Change. Texas Tech
University Press, Lubbock, xiv + 534 pp.

Schmidly, D. J. 2004. The Mammals of Texas. University of Texas Press, Austin, xviii
+ 501 pp.

Sierra Club. 2006. Sierra Club Conservation Policies: Feral Burro Management Policy.
http://www.sierraclub.org/policy/conservation/feral.asp

Stangl, F. B., Jr., W. W. Dalquest & R. R. Hollander. 1994. Evolution of a desert
mammalian fauna: a 10,000-year history of mammals from Culberson and Jeff Davis
counties, Trans-Pecos Texas. Midwestern State University Press, Wichita Falls,
Texas, xix + 264 pp.

Taylor, W. P., & W. B. Davis. 1947. The Mammals of Texas. Texas Game, Fish &
Oyster Commission, Austin, Bulletin 27:1-79.

Woodward, S. E., & R. D. Ohmart. 1976. Habitat use and fecal analysis of feral burros
{Equus asinus) in the Chemehuevi Mountains, California, 1974. Journal of Range
Management, 29:482-485.

Yancey, F. D., IF 1997. The mammals of Big Bend Ranch State Park, Texas. Special
Publications, Museum of Texas Tech University, 39:1-210.

FBS at: frederick.stangl@mwsu.edu

TEXAS J. OF SCI. 59(1):1 1-22

FEBRUARY, 2007

HABITAT ATTRIBUTES AND POPULATION SIZE OF
TEXAS KANGAROO RATS ON AN INTENSELY GRAZED PASTURE
IN WICHITA COUNTY, TEXAS

Jim R. Goetze*, William C. Stasey, Allan D. Nelson
and Philip D. Sudman

^Science Department, Laredo Community College, Laredo, Texas 78040 and
Department of Biology, Tarleton State University, Box T-OlOO
Stephenville, Texas 76402

Abstract.-An assessment of burrows was conducted at a 15-ha site in north-central
Texas known to contain a population of Texas kangaroo rats {Dipodomys elator). The
greatest numbers of burrows were in loose, elevated soils where 30-year-old unbumed
brush piles had decayed, followed by elevated, open areas. The least occupied habitats
were fence rows, woody vegetation such as lotebush {Zizyphus obtusifolia) and honey
mesquite {Prosopis glandulosa), and rocks. Texas kangaroo rats favored slightly
elevated, well-drained, clay loam soils. Dominant vegetation at the study site was
indicative of disturbances such as intense grazing by livestock. Burrows were associated
with high percentages of bare ground, and herbaceous and woody vegetation of low
height. Vegetation in areas immediately surrounding burrows was significantly different
from that along transects leading away from burrows in percentages of bare ground and
grasses. Dipodomys elator and four other mammalian species were captured during 640
trap nights. The population size of D. elator was estimated to be 33 (± 6) individuals.
Sixty-six burrows were counted and multiple burrow use by the Texas kangaroo rat is
hypothesized. Counting of burrows is less expensive and time consuming than trapping,
and if suitable conversion factors can be established by additional investigations, burrow
counts have potential as a means of estimating size of populations that would provide
information useful to conservation of this endemic, state-listed, threatened species.

The Texas kangaroo rat {Dipodomys elator) is listed as a
threatened species by the Texas Parks and Wildlife Department
(Schmidly 2004). Reasons for D. elator being listed are largely based
on its apparent scarcity and small geographic range. The International
Union for Conservation of Nature (1986) lists habitat loss and
degradation resulting from expanding agricultural and infrastructure
development as the major threats to continued existence of D. elator.

An association between honey mesquite {Prosopis glandulosa) and
the Texas kangaroo rat has been well documented (Bailey 1905; Blair
1954; Carter et al. 1985; Chapman 1972; Dalquest & Collier 1964;
Martin & Matocha 1972, 1991; Roberts & Packard 1973; Schmidly
2004). More recently Stangl et al. (1992) noted that extensive stands

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

of mesquite within the range of D. elator may not be as critical as
once believed for survival and persistence of the Texas kangaroo rat.
Also, while working at the Wichita County study site and another
nearby site, Stangl et al. (1992) discussed opportunistic use of habitat
affected by human disturbances. However, during that investigation,
characteristics of burrows and size of populations was not quantified.
Obtaining information on habitat characteristics critical to species
survival and documenting demographic changes of D. elator
populations must be accomplished to evaluate the conservation status
of the Texas kangaroo rat (Jones et al. 1988). Therefore, the purposes
of this study were to locate potential burrows of D. elator within this
study site, relate associations of burrows with soils and human
disturbance, quantify vegetation associated with burrows, and
examine whether there is a correlation between number of burrows
and size of population.

Methods

The study site, located in Wichita County, Texas, 3 km N Lake
Buffalo Creek Reservoir (location in decimal degrees; 34.02653 N,
98.75991 W), was previously described by Stangl et al. (1992) as
study site 2 - Goetze property. The grazing regimen of this site has
remained virtually unchanged since described by Stangl et al. (1992).
The 15-ha area is fenced and bordered by wheat fields on its eastern
and western sides and contains constructed earthen ponds and erosion
features, such as rills and gullies. There is additional habitat occupied
by Texas kangaroo rats to the north and south of this study site and it
probably is open to migration from these areas.

Beginning in 1975, the study site was mechanically cleared of
brush and subsequently sprayed with an herbicide four times over five
year intervals. This brush was piled into mounds scattered throughout
this pasture (Stangl et al. 1992), but was left unbumed. Over a period
of 30 years, the brush decayed and collected soil to form low earthen
mounds about 0.5 m in height. A few dead stumps still persist in most
of these elevated brush piles. As a result of mechanical and chemical
treatments, brush at the site is short (1-2 m in height) but still
relatively dense. A count of 54 honey mesquites, two lotebushes, and

GOETZE ET AL.

two net-leaf hackberry {Celtis laevigata var. reticulata) in a one-ha
quadrat was considered typical of the site. About 30 large, sandstone
rocks are associated with constructed earthen dams surrounding the
three ponds within the study site.

The study site was surveyed for burrows of the Texas kangaroo rat
by the investigators walking about 3 m apart over the entire area. The
survey occurred on 18-26 May 2005. Burrows belonging to D. elator
were identified based on diameter and orientation of entrance/exit hole
(see Fig. 3 in Stangl et al. 1992). Stasey (2005) noted a significant
difference between size of burrow entrance and angle of entry to the
burrow between D. elator and other rodents of similar size. Distinct
trails and dust-bathing areas led away from these burrows and these
runways sometimes connected to other distant burrows. The specific
location of each burrow was recorded in decimal degrees using a
Garmin GPS- 12 unit and were categorized as being associated with
human-mediated disturbances such as old brush piles, fence rows, and
rocks associated with constructed earthen dams, or other available
habitats that contained elevated, open areas, lotebushes, or honey
mesquites.

A geo-referenced base map was produced with a Manifold 5.0
(Manifold System Ltd, 2003) GIS system using a 1-m resolution,
digital orthophoto quadrangle obtained from IntraSearch, Denver,
Colorado. This digital map was loaded into a Manifold 5.50 GIS
system. Spatial and habitat data were entered into data tables and
imported onto the base map as layers. A geo-referenced drawing of
the major soil series of the study site was added to the layered map to
ascertain whether a specific type of soil was most favored by the
Texas kangaroo rat. Information about soils was obtained from the
soil survey of Wichita County, Texas (Richardson et al. 1977). Area
of the study site and the extent of soil coverage (in m^) were obtained
using the Tracker function in the Manifold software.

Vegetation was sampled during May 2005. A 1-m^ quadrat made
from PVC pipe was placed directly over 10 burrows where D. elator
was captured. Within each quadrat, vegetative richness was recorded

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

as total number of species present (Brower et al. 1990). Percentage
coverage of grass, forbs, bare ground, woody vegetation, and rocks or
stumps within each quadrat were recorded, as was average herbaceous
vegetation height (obtained by averaging the height of the herbaceous
vegetation 15 cm interior to each comer of the quadrat). If woody
vegetation was present, its height also was recorded. To quantify
vegetation of habitat away from burrows, 20 m north-to- south and
east-to-west transects, bisecting at the burrow entrance, were
evaluated. Vegetation type (grass, forb, bare ground, woody, or
other), height, and vertical intercept (distance from ground to woody
species in cm directly over meter-point) was recorded at each meter-
point for a total of 40 data points for each burrow (Brower et al.
1990). Specimens of the dominant herbaceous and woody plants were
collected and placed in a plant press. These vegetation vouchers were
deposited in the herbarium of Tarleton State University (TAG).

Ordinal vegetation percentage data between quadrats and transects
were compared within the study site with the Wilcoxon Mann-
Whitney test (SAS Institute 1999). A paired ^test (SAS Institute
1999) was used to compare vegetational height. Herbaceous height
and percentages of grasses, forbs, bare ground, woody, and other
categories were all evaluated for statistically significant (P<0.05)
differences.

To test accuracy of counting burrows to estimate population size,
trapping was conducted by placing three 7.5 by 8.8 by 30 cm Sherman
Live Traps within 0.10 to 0.50 m of each burrow entrance, with the
open end of each trap facing the entrance (Cross & Waser 2000).
Traps were baited with dry oatmeal. Trapping was conducted during
these dates in 2005 (parentheses indicate number of trap nights): May
18-24 (178), June 21-25 (210), July 6-7 (84), and July 19-22 (168).

Captured animals were tagged with passive integrated transponders
(PIT tags) to determine rates of recapture for specific individuals. PIT
tags were implanted subcutaneously immediately posterior to the
cranium by means of a syringe. To minimize handling time and other
stresses, anesthesia was not used (Schooley et al. 1993). Syringes

GOETZE ET AL.

were sterilized between implantations with 91% aleohol. Each
animal's total length, gender, and reproductive condition were
recorded at the time of capture. As a result of trap mortality, one
individual was prepared as a skin and skull voucher specimen (TSU
1294) on 20 May 2005. Program MARK was used to estimate size of
population from the trapping data (White & Burnham 1997) with the
integrated POP AN program using the Jolly-Seber algorithm and
assuming an open population with equal probabilities of capture and
survival.

Results

Of the two major soil associations within the study area, most D.
elator burrows (56) were in Kamay silt loam soils (9.8 ha) and the
remaining 10 burrows occurred in Asa-Portales soils (5.2 ha).
Twenty-one of the 22 Texas kangaroo rats were captured in the
Kamay soil association.

Of the vegetation sampled within quadrats, mean percentage cover
of grasses was 25.1 (range 1 - 55) and little barley {Hordium pusillum)
was always the dominant grass (Table 1). Mean percentage cover of
forbs was 17 (range 1 - 35) and most quadrats contained Virginia
pepperweed (Lepidium virginicum). Other herbaceous dominants
included common broomweed {Gutierrezia dracunculoides), hog
potato {Hoffmanns eggia glauca), and western ragweed {Ambrosia
psilostachya). Mean percentage cover of woody vegetation was six
(range 0 - 50), evenly distributed between lotebush and honey
mesquite. Mean percentage of bare ground was 49.9 (range 0 - 80)
and stumps comprised a minor component of the habitat. Mean
percentage richness was 5.8 (range 3-10). Mean herbaceous height
was 7.1 cm (range 2-40 cm) and mean woody height was 15.9 cm
(range 20 - 121 cm).

There was significantly more bare ground within quadrats (49.9%)
than along transects (22.3%; Table 2). Likewise, there was
significantly less grass within quadrats (25.1%) than along transects
(54.3%). There were no significant differences between other
compared parameters (Table 1).

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

Table 1. Mean percentage cover of bare ground, forbs, grasses, woody vegetation, and
stumps as well as average herbaceous and woody vegetation heights compared
between 10 quadrats and 40 transects in May 2005. Percentage data were evaluated
by a Wilcoxon Mann-Whitney test. Height data were compared by a paired t-test.
Standard deviations for data are indicated by parentheses and significant differences
are denoted by an asterisk.

Parameter

Quadrats

Transects

% Bare Ground

49.9 (± 24.0)

22.3 (± 8.7)

0.010*

% Forbs

17.0 (± 12.9)

22.9 (± 9.8)

0.177

% Grasses

25.1 (± 18.9)

54.3 (± 9.5)

0.002*

% Woody

6.0 (± 15.8)

0.5 (± 1.6)

0.256

% Stump

2.0 (± 4.2)

0.0(± 0.0)

0.092

Herbaceous

Height (cm)

7.1 (± 6.7)

7.2 (± 3.7)

0.934

Woody

Height (cm)

15.9 (±38.8)

10.2 (± 32.3)

0.700

Sixty-six active burrows were found within the study site and the
burrow associations were as follows: Four active burrows were found
at or near the bases of honey mesquite trees, and five active burrows
were underneath large, sandstone rocks. Six active burrows were
underneath lotebushes. Seven active burrows had been constructed in
the elevated soils associated with fence rows. Nineteen aetive
burrows were found in elevated, open areas, and soils assoeiated with
30 year old brush piles contained the greatest number (25) active
burrows.

A total of 640 trap nights was conducted within the sampling area.
Forty-five eaptures of D. elator, representing 18 individuals, were
obtained. Ten Chaetodipus hispidus were captured over the course of
the trapping period. Also, six Spermophilus tridecemlineatus , four
Neotoma micropus, and two Peromyscus leucopus were captured.
The 18 Texas kangaroo rats were captured at 22 different burrows.
Habitat associations of the 22 burrows where D. elator was captured
were as follows: mesquite (1), rock (1), lotebush (2), brush piles (8),
and prairie mounds (10).

GOETZE ET AL.

Utilizing the MARK program, a population estimate of 33 ± 6
individuals was obtained for the study area. Ninety-five pereent
confidence limits ranged from 25 - 49 individuals. Of the 18 D. elator
captured, eight were caught at multiple burrows. Of those eight
individuals, five were captured at two different burrows, and three
were captured at three different burrows. Of the 22 burrows where D.
elator was captured, multiple individuals were captured at five
burrows. Multiple D. elator were caught at the same burrow on the
same trap night only twice. On a single occasion, a male and female
where captured together at the same burrow.

Discussion

5'ozA.-Both major soil associations within the study site are
categorized as clay loams. The Kamay soils are more favored by D.
elator and dominate throughout most of the study site. Kamay soils
are gently sloping and were formed over ancient alluvium deposits
from red-bed clay and shale. These soils are well-drained but slowly
permeable. The shrink-swell potential in some areas may be severe
because of high underlying clay content (Richardson et al. 1977).
Asa-Portales soils formed in alluvial materials of recent age and occur
in creeks and intermittent streams (Richardson et al. 1977). Because
Asa-Portales soils are subject to frequent flooding, they provide only
marginal habitat for Texas kangaroo rats. Although 15% of burrows
occurred in this soil association, most of the burrows were found
along the margins of the Asa-Portales soils.

Packard & Roberts (1973) noted that D. elator was not in sandy
soils of the Red River terraces within Wichita County north of the
study site. However, Martin & Matocha (1991) indicated that clay or
clay loam soils may not always be used for burrow sites because a few
D. elator have been found in soils with high sand content.

Kamay soils at the study site are classified as alfisols. Alfisols of
the Blackland Prairie region of North Central Texas contain soil
features known as prairie mounds (Diggs et al. 1999). These mounds
were formed as a result of shrink-swell and soil overturn properties of
clays underlying alfisols. Although raised, open areas are of small

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

elevation (about 20 cm or less) at the study site, the mounds appear to
provide favorable burrow sites for D. elator as they comprised 29% of
the total habitat associations. Thus, prairie mounds and other
naturally occurring habitat heterogeneity features likely played
important roles in determining distribution patterns of D. elator within
Wichita County. Stangl et al. (1992) proposed that past activity of
Bison bison may have created suitable habitat for D. elator by creating
disturbances such as wallows and by intense grazing. However,
because of the migratory nature of bison and its large range, it seems
unlikely that bison alone could have created and maintained suitable
habitat for Texas kangaroo rats. It is proposed that natural habitat
heterogeneity, due in part to soil properties that form prairie mounds,
also may be an important factor in determining the past and present
distribution of the Texas kangaroo rat.

Vegetation. -LiUIq barley, common broomweed, hog potato,
Virginia pepperweed, and western ragweed occur in disturbed
habitats, and common broomweed is often an indicator of heavy
grazing (Diggs et al. 1999). These plants were dominant species
associated with burrows of Texas kangaroo rats (Table 1) and their
occurrence is likely caused by intense grazing by cattle and rodent
activity around the burrows. Habitat of D. elator was dominated by
short, herbaceous vegetation (2.0 - 40.0 cm in height) with little
overhead woody cover, and there was a significantly greater amount
of bare ground and less grass within quadrats as compared to transects
leading away from the burrow. There is general agreement that D.
elator requires a sparse, short-grassland habitat (Carter et al. 1985;
Dalquest & Collier 1964; Roberts & Packard 1973; Stangl et al.
1992), and findings from this current study support this conclusion,
although as discussed earlier, there has been disagreement concerning
the importance of mesquite within habitats.

At present, habitat as described above usually is not typical for
Wichita County, Texas. Nearly all tillable land is under cultivation
within the range of D. elator. Routine tilling and resulting
monocultures of these fields render these areas uninhabitable for
Texas kangaroo rats (Stangl et al. 1992). Areas not in crop production

GOETZE ET AL.

are developed for gas and oil exploration or used as rangeland.
Associated disturbances, such as road construction, and discarded
equipment that accumulates soil are opportunistically used by Texas
kangaroo rats (Roberts & Packard 1973; Stangl et al. 1992).

The use of fire to control woody species is precluded by presence
of oil field equipment, and costs of mechanical brush control often are
prohibitive. These circumstances may allow areas to develop dense
stands of mature honey mesquite, wherein the herbaceous vegetation
becomes tall and dense. In 1985, D. elator was recorded for two
separate locations in Hardeman County, Texas. When the sites were
visited in 1990, the vegetation had become much denser and D. elator
had been extirpated from the locations (Stangl et al. 1992). The
Goetze property is a location of intensive grazing by cattle and the
population of D. elator is known to have persisted at this site since at
least 1930 (Oscar and Ernest Goetze, pers. comm.). Intensive grazing
for maintenance of D. elator habitat has been documented (Chapman
1972; Stangl et al. 1992) and the grazing regime at the study site may
be a factor accounting for the long-term persistence of the population.

Burrow associations -Tht greatest number of burrows of the
Texas kangaroo rat were associated with loose, elevated soils where
30-year-old, unbumed brush piles had decayed {n = 25). Burrows
also were found in fence rows (7) adjacent to wheat fields located on
the east and west sides of the study site and under large rocks
occurring in earthen dams (5). This is in agreement with Stangl et al.
(1992), who concluded that D. elator opportunistically used disturbed
habitats including brush piles and fence rows.

Second in abundance were associations with elevated, open areas
{n = 19), some of which could have formed due to disturbance during
the clearing of brush whereas others may be prairie mounds formed by
alternating cycles of soil shrinkage and swell to form prairie mounds.
Honey mesquite and lotebush associations were less abundant (4 and
6 burrows, respectively). Many researchers have emphasized an
affinity between D. elator and mesquite (Bailey 1905; Blair 1954;
Carter et al. 1985; Chapman 1972; Dalquest & Collier 1964; Martin &

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

Matocha 1972, 1991; Roberts & Packard 1973; Schmidly 2004).
Although a high number of small honey mesquite (54/ha) was present
on the study area, it was found that loose, elevated soils formed from
decay of old brush piles and elevated, open areas were the most
important habitats at this site. Additionally, lotebushes were of
greater importance (based on this shrub’s overall lower density and
greater number of burrow associations) as a habitat feature at the site
than honey mesquites.

Burrow use and population estimates -It is suggested that D.
elator may occupy more than one burrow. Forty-four percent of the
captured animals possibly used more than one burrow. Upon release,
these individuals always returned to the nearest burrow. On two
occasions, two individuals were captured at the same burrow entrance.
Runways were observed connecting burrows and, by using night-
vision scopes, animals were observed leaving one burrow and
traveling along the runway to enter an adjacent burrow. Therefore,
the current results differ from those of Packard & Roberts (1973), who
reported rare use of multiple burrows.

The population estimate of D. elator for this site was 33 ± 6
individuals. Sixty-six burrows were mapped and preliminary observa¬
tions suggest multiple burrow use. Based upon the data and
observations, D. elator appears to use two or more burrows per
individual at this particular location. Intensive trapping (18 periods
over 9.5 weeks) was conducted to obtain a reliable estimate of
population size. However, intensive trapping of localities where D.
elator occurs to estimate population sizes over a large geographic
range seems impractical due to private ownership of most lands and
fragmented habitats within this species' range. If a suitable
conversion factor for number of burrows per animal can be obtained
by additional studies, the burrow counting method could prove to be
much quicker, less labor intensive, and less disruptive for D. elator
populations than trapping.

By dividing the area of the study site (15 ha) by the population
estimate (33), one obtains estimates of population densities of one D.

GOETZE ET AL.

elator /0.45 ha (or approximately 2/ha). This estimated density is
much lower than the previously reported 10/ha for the same general
area in Wichita County, Texas (Roberts & Packard 1973). This
discrepancy is difficult to explain. Two possible reasons are (1) either
the sites where Roberts & Packard (1973) conducted their research
were more conducive to high populations or (2) overall population
densities have declined since 1973. To determine if populations are
indeed declining, many more population estimates need to be obtained
throughout the range of the Texas kangaroo rat. Burrow mapping
surveys might provide a cost-effective means of accomplishing this
objective. Burrow counts and trapping to estimate population sizes
have been employed of studies of the banner-tailed kangaroo rat, D.
spectabilis, (Cross & Waser 2000) and Stephens' kangaroo rat, D.
stephensi, (Brock & Kelt 2004). Results varied according to locality
with D. stephensi, whereas burrow trapping was found to be a reliable
method in censusing D. spectabilis. However, additional studies are
needed to determine how accurately burrow counting methods may
estimate D. elator population numbers.

Acknowledgments

We thank Mr. and Mrs. Oscar Goetze and Mr. Ernest Goetze for
allowing us access to their properties within Wichita County. This
study could not have been completed without their kind permission.
We acknowledge R. Wittie and B. Lambert for assistance with
statistical analyses and a number of reviewers for improving the
paper. We thank Mike Miller and Texas Parks and Wildlife for
providing PIT tags and associated equipment. This study was
conducted under Texas Parks and Wildlife permit SPR-0496-775.

Literature Cited

Bailey, V. 1905. Biological survey of Texas. North American Fauna, 24:1-222.

Blair, W. F. 1954. Mammals of the Mesquite Plains Biotic District in Texas and
Oklahoma and speciation in the central grasslands. Texas J. of Sci., 6(3):235-264.
Brock, R. E & D. A. Kelt. 2004. Conservation and social structure of Stephens'
kangaroo rat: Implications from burrow-use behavior. J. Mamm., 85(1): 51-57.
Brower, J. E., J. H. Zar & C. N. vonEnde. 1990. Field and laboratory methods for
general ecology. Wm. C. Brown Publ., Dubuque, lA. 237 pp.

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

Carter, D. C., W. D. Webster, J. K. Jones, Jr., C. Jones & R. D. Suttkus. 1985.
Dipodomys elator. Mammalian Species 232:1-3.

Chapman, B. R. 1972. Food habits of Loring's kangaroo rat, Dipodomys elator. J.
Mamm., 53(4):877-880.

Cross, C. L. & P. M. Waser. 2000. Estimating population size in the banner-tailed
kangaroo rat. Southwestern Nat., 45(2): 176- 183.

Dalquest, W. W. & G. Collier. 1964. Notes on Dipodomys elator, a rare kangaroo rat.
Southwestern Nat., 9(3): 146- 150.

Diggs, G. M., Jr., B. L. Lipscomb & R. J. O'Kennon. 1999. Shinner and Mahler's
illustrated flora of North Central Texas. Botanical Research Institute of Texas, Ft.
Worth, Texas, 1626 pp.

International Union for Conservation of Nature and Natural Resources. 1986. lUCN red
list of threatened animals. lUCN, Cambridge, U. K. 105 pp.

Jones, C., M. A. Bogan & L. M. Mount. 1988. Status of the Texas kangaroo rat
{Dipodomys elator). Texas J. of Sci., 40(3):249-258.

Manifold Net Ltd. 2003. Manifold System 5.50 SP2. CDA International Ltd. Carson
City, Nevada.

Martin, R. E. & K. G. Matocha. 1972. Distributional status of the kangaroo rat
Dipodomys elator. J. Mamm., 53(4):873-877.

Martin, R. E. & K. G. Matocha. 1991. The Texas kangaroo rat, Dipodomys elator, from
Motley County, Texas, with notes on habitat attributes. Southwestern Nat.,
36(3):354-356.

Packard, R. L. & J. D. Roberts. 1973. Observations on the behavior of the Texas
kangaroo rat. Mammalia, 55(4):680-682.

Richardson, W. E., A. R. Goerdel & K. T. Lofton. 1977. Soil survey of Wichita County,
Texas. United States Department of Agriculture Soil Conservation Service, 88pp.

Roberts, J. D. & R. L. Packard. 1973. Comments on movements, home range and
ecology of the Texas kangaroo rat, Dipodomys elator Merriam. J. Mamm., 54(4):
957-962.

SAS Institute. 1999. SAS companion for the Microsoft windows environment, version
8. North Carolina SAS Publishing. Cary, North Carolina.

Schmidly, D. J. 2004. The mammals of Texas, revised edition. University of Texas
Press, Austin, 501 pp.

Schooley, R. L., B. Van Home & K. P. Burnham. 1993. Passive integrated transponders
for marking free-ranging Townsend’s ground squirrels. J. Mamm., 74(2):480-484.

Stangl, F. B., Jr., T. S. Schafer, J. R. Goetze & W. Pinchak. 1992. Opportunistic use of
modified and disturbed habitat by the Texas kangaroo rat {Dipodomys elator). Texas
J. ofSci.,44(l):25-35.

Stasey, W. C. 2005. An evaluation of Texas kangaroo rat {Dipodomys elator):
Biological habits and population estimation. Unpubl. Masters Thesis. Tarleton State
University, 45 pp.

White, G. C. & K. P. Burnham. 1997. Program MARK: survival estimation from
populations of marked animals. Colorado Cooperative Fish and Wildlife Research
Unit, Colorado State University, 33 pp.

JRG at: jgoetze@laredo.edu

TEXAS J. SCI. 59(l):23-32

FEBRUARY, 2007

INFESTATION BY CHIGGER MITES IN TWO LIZARD SPECIES
FROM A DUNE HABITAT OF NORTHERN MEXICO

Cristina Garcia-De la Pena, Gamaliel Castaneda
and Cameron W. Barrows*

Facultad de Ciencias Bioldgicas, Universidad Autonoma de Nuevo Leon
C.P. 66450, San Nicolas de los Garza, Nuevo Leon, Mexico
* Center for Conservation Biology, University of California
75-080 Frank Sinatra Drive, Palm Desert, California 92211, USA

Abstract -The occurrence and levels of infestation by larvae of the chigger
mite Eutrombicula alfreddugesi on the desert lizards Uma exsul and Uta
stejnegeri were studied on the sand dunes of Viesca, Coahuila, Mexico. Chiggers
were observed on 68.1% of U. exsul males with an average of 7.81 ± 2.7 chiggers
per lizard; chiggers occurred on 54.5% of the females and averaged 4.18 ± 1.5
mites per lizard. For U. stejnegeri males, chiggers were observed on 100% of the
captured lizards and each lizard averaged 61.12 ± 6.8 chiggers; 86.8% of the
females were parasitized and they averaged 38.05 ± 6.8 mites per lizard.
Regression analyses were used to test the relationship between chigger infestation
levels and lizard body size; however, no relationships were detected. All chiggers
were found on the neck folds in both lizard species. Morphology (shallow neck
folds), microhabitat selection (sandy and sunny), and burrowing behavior could
influence the infestation pattern of U. exsul. Home range size (for males), deeper
neck folds and selection for cooler and shaded microhabitats with higher relative
humidity could account for the greater chigger infestation levels found on U.
stejnegeri.

Resumen.-Se estudid la prevalencia e intensidad de infestacion por larvas de
Eutrombicula alfreddugesi en los saurios deserticos Uma exsul y Uta stejnegeri
en las dunas de arena de Viesca, Coahuila, Mexico. En U. exsul la prevalencia e
infestacion en los machos fue de 68.1% y 7.81 ± 2.7 acaros respectivamente, y en
las hembras fue de 54.5% y 4.18 ± 1.5 acaros. En U. stejnegeri fue del 100% y
61.12 ± 6.8 acaros para los machos, y 86.8% y 38.05 ± 6.8 acaros para las
hembras. Utilizando el analisis de regresion se determine que no hubo relacion
entre la intensidad de infestacion y el tamano de los saurios. Todos los acaros se
encontraron dentro de los pliegues del cuello en ambas especies. La morfologia
(pliegues poco profimdos en el cuello), la seleccion de microhabitat (arenoso y
soleado) y la conducta de entierro son caracteristicas que pueden influenciar el
patrdn de infestacion de U. exsul. El tamano del ambito hogareno (en los
machos), los pliegues profimdos en el cuello y una seleccion de microhabitats
templados y sombreados con una mayor humedad relativa, pueden intervenir en el
alto grado de infestacidn de U. stejnegeri.

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

Chigger mite occurrence and parasitism rates on lizards have
been associated with host species’ characteristics such as
morphology, behavior, and gender. Size of the lizard, scale type,
and the presence of folds or pockets, are morphological variables
that affect chigger parasitism (Cunha-Barros & Rocha 2000;
Salvador et al. 1999). The foraging mode and the activity
patterns are examples of behavioral factors (Clopton & Gold
1993; Cunha-Barros et al. 2003), and home range size and male
testosterone levels (Salvador et al. 1996; Talleklint-Eisen &
Eisen 1999) are examples of gender-specific attributes that
influence chigger mite parasitism on lizards.

Eutrombicula alfreddugesi is a chigger mite with a wide
distribution. It has been reported throughout the American
continent from southern Canada to Argentina and the Caribbean
Islands (Loomis & Wrenn 1984; Lareschl et al. 2003; Daniel &
StekoTnikov 2004). Larval chiggers attach themselves to the
skin of amphibians, reptiles, birds and mammals (Daniel &
StekoTnikov 2004), and have medical importance because they
cause dermatitis in humans (Potts 2001). The parasitic
occurrence of this mite on lizards has been studied mainly in the
U.S. (Klukowski 2004) and in South America (Cunha-Barros et
al. 2003). In Mexico, the few mite/llzard studies that have been
conducted were mainly in forest habitats (Garcia-De la Pena et al.
2005a; 2005b); mite/lizard relationships in desert habitats have
not been previously studied.

In the sand dunes of Viesca, Coahuila, the Coahuila fringe¬
toed lizard Uma exsul (currently under a special protection
category [SEMARNAT 2001]) and the side blotched lizard Uta
stejnegeri are sympatric species. Because sympatric lizards can
have species-specific levels of parasitism related to their
morphological, behavioral and gender specific characteristics
(Cunha-Barros & Rocha 2000), determining the occurrence and
infestation levels by E. alfreddugesi on these two phrynosomatid

GARCIA-DE LA PENA, CASTANEDA & BARROWS

species is important. In this study the proportion of lizards
parasitized by at least one mite, the intensity of infestation, its
relationship with the size and gender of the lizards, and the
distribution of E. alfreddugesi over the body was calculated for
both lizard species.

Materials and Methods

The study area was located in the middle of Chihuahuan
Desert, to the southwest of Coahuila State, Mexico, municipality
ofViesca(25^26’27”N, 102°55’15”W) (Fig. 1). This region has
an elevation of 1 100 m. The annual average precipitation is 250
mm, occurring mainly between July to September (INEGI 1988).
Annual temperature averages 21° C, with December and January
being the coldest months, and July and August the warmest
(Garcia 2004). Perennial dune vegetation consists of creosote
bush (Larrea tridentatd), black seepweed {Suaeda nigrescens),
and honey mesquite (Prosopis glandulosa) (Rzedowski 1978). In
November there are high densities of annual plants including
desert marigold (Baileya multiradiata) and woolly tidestromia
{Tidestromia lanuginosa) (Garcia-De la Pena et al. 2007).

Data collection was limited to November 2004 to factor out
possible seasonal effects on chigger infestation patterns.
Individuals of U. exsul and U. stejnegeri were captured with a
noose or by hand. Data collected for each lizard included: sex
(hemipenal eversion used to identify U. exsul males, and dorsal
pattern and coloration to identify U. stejnegeri males), snout-vent
length (SVL, to the nearest 0.1 mm), weight (W) measured to the
nearest 0.1 g with a 30-g Pesola™ spring scale, and total number
of chigger mites carried (Tm). To measure T^, each specimen
was carefully examined with special attention to neck folds,
axillae and posfemoral pleats. The mites were removed in the
field using wetted cotton swabs. The red color of the mites
facilitated counting them on the surface of the cotton with a
magnifying glass and then collecting them. Each lizard was

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

Fig. 1. Map of study area.

temporally marked on its back using an indelible marker to avoid
recaptures. Lizards were released at the place of capture.

Kolmogorov-Smimov goodness of fit tests revealed SVL, W,
and Tm to be normality distributed, so t-tests were used to
compare means of SVL, W, and Tm between sexes of each
species, and between species (combined data for females and
males). To identify differences in infestation intensity between
sexes for each species and between species. Analyses of
Covariance (SVL and weight as covariates) were used.
Regression analyses were also conducted between SVL and T^,
and W and for both species. All tests assumed to be
significant at a = 0.05. Measurements are reported as mean ± SE.

Results

Mite parasitism levels for 44 Uma exsul (22 males and 22
females) and 88 Uta stejnegeri (50 males and 38 females) were
measured. Results revealed that 68.1% of males and 54.5% of
females of U. exsul showed infestation. For U. stejnegeri, 100%

GARCIA-DE LA PENA, CASTANEDA & BARROWS

of males and 86.8% of females carried at least one chigger. The
difference in SVL of males and females in U. exsul did not reach
traditional levels of statisitical significance {t = 1.91, d.f. = 42, P
= 0.06); however, males averaged statistically heavier than
females {t = 2.28, d.f. = 42, P = 0.02) (Table 1). The mean SVL
of male U. stejnegeri was significantly greater {t = 8.48, d.f. = 86,
P = 0.0001) and the males were heavier than females {t = 7.87,
d.f. = ?>6, P = 0.0001) (Table 1). As with SVL, differences in
mean T^ values for U. exsul males and females (Table 1) did not
reach traditional levels of statistical significance {ANCOVA: F =
3.04, d.f. = 1.40, P = 0.08). Males of U. stejnegeri did have
greater mite infestations than females (ANCOVA: F = 5.92, d.f. =
1.84, P- 0.01) (Table 1).

There was no relationship between SVL and T^ in either
species (^7. exsul: r = 0.04, F = 2.09, d.f. = 1.42, P = 0.15; U.
stejnegeri: f = 0.04, F = 2.02, d.f. = 1.42, P = 0.16) or between
W and fj. exsul: r = 0.006, F = 0.48, df = 1 .86, P = 0.48; U.
stejnegeri: f = O.W, F = 1.07, d.f. = 1.86, P = 0.30). Ignoring
gender differences, U. exsul averaged larger than U. stejnegeri
[SVL: U. exsul (61.59 ± 1.8 mm), U. stejnegeri (48.84 ± 0.4
mm), / = 8.63, df = 130, P = 0.0001; Weight: U. exsul (8.72 ±
0.9 g), U. stejnegeri (4.78 ±0.1 g), / = 5.49, d.f. = 130, P =
0.0001. The overall intensity of mite infestations on U.
stejnegeri (51.15 ± 4.9 chiggers, range = 0-218) was greater than
on U. exsul (6 ± 1.5 chiggers, range = 0-56), ANCOVA: F =
18.93, d.f. = 1.128, P = 0.0001]. Mites were found exclusively in
the neck folds of males and females of both species.

Discussion

For both sexes of Uma exsul and Uta stejnegeri chiggers
occurred on more than 50% of the individuals, indicating that
Eutrombieula alfreddugesi did not appear to discriminate
between the two lizard species as hosts. The level of mite
infestation in males and females of U. exsul was similar and was

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

Table 1. Snouth-vent length (SVL), weight (W), total number of chigger mites carried
(Tm) and Tm range for males and females of Uma exsul and Uta stejnegeri.

Species

Sex

Uma exsul

Uta stejnegeri

SVL

65.09 ±3.4 mm

58.09 ± 1.2 mm

51.36 ± 0.4 mm

45.52 ±3.5 mm

10.84 ± 1.8 g

6.60 ± 0.4 g

5.55±0.1 g

3.76 ±0.1 g

T.,

7.81 ±2.7

4.18± 1.5

61.12±6.8

38.05 ±6.8

Tm range

0-56

0-31

3-218

0-188

not related to the SVL or W of the individuals. In eontrast, the
number of chiggers found on males of U. stejnegeri was higher
than on females. The size of individuals of U. stejnegeri did not
influenee the level of infestation, however, it may be that the
home range of this speeies is related to the number of ehiggers
that parasite it. It was observed that during the non-breeding fall
season (November), the males of U. stejnegeri oeeupied home
ranges twiee the size as those of the females in eomparison with
the breeding season. When males eover a greater area, they
eneounter many mierohabitats. This inereases the probability of
eontacting more ehiggers and beeoming parasitized by them
(Davis & Ford 1983; Talleklint-Eisen & Eisen 1999).

Individuals of U. exsul had redueed ehigger levels eompared
to U. stejnegeri. The mierohabitats whieh eaeh species occupies
may explain this finding. Clopton & Gold (1993) observed that
Eutrombicula alfreddugesi prefers areas of low to moderate
temperature, high relative humidity, little incidence of sunlight,
and dense vegetation. Garcia-De la Pena et al. (2007) observed
that U. exsul and U. stejnegeri select different mierohabitats on
the dunes of Viesca. U. exsul primarily use creosote bushes for
thermoregulating and foraging. Creosote has an open canopy that
allows considerable sunlight to reach the ground below, and
generally grows in very sandy areas where rodent burrow

GARCIA-DE LA PENA, CASTANEDA & BARROWS

abundance is low. During this study, air and substrate
temperatures around 30® C and 38® C, respectively, and a relative
humidity around 20% were observed in these areas. This may
explain the low numbers of chiggers in this microhabitat, which
would explain the relatively low mite levels found on U. exsuL
In the drier and hotter sand dunes in the northwest portion of the
Sonoran Desert, another fringe-toed lizard, Uma inornata, was
never observed with mite infestations (Barrows, pers. comm.).
Also, the sand burrowing behavior that characterizes members of
the genus Uma may offer an additional explanation for the fewer
chiggers found on U. exsul. Fringe-toed lizards dive into loose
sand and can ‘‘swim” beneath the sand surface for short distances
to avoid thermal stress and to escape predators (Stebbins 1944).
This behavior results in friction of their skin against the sand,
which can act like an abrasive that removes the mites that are not
well protected by the folds of the neck.

In contrast, U. stejnegeri has been more frequently observed
thermoregulating and foraging beneath black seepweed (Garcia-
De la Pena et al. 2007). This shrub has a dense canopy and
usually grows in compact sand mounds. Its roots and the
burrows that rodents construct underneath are likely refuges and
nesting sites for this lizard. Air and substrate temperatures
around 28® C and 35® C, respectively, and a relative humidity
around 25% were observed in these areas. Because the
microclimates that exist underneath this plant and within the
burrows are cooler and more humid than on the more open sand
dunes, the abundance of chiggers would be greater and so would
explain the higher chigger infestations on U. stejnegeri.

In both lizard species, mites were found exclusively in the
neck folds. Chiggers infesting U. stejnegeri were concentrated in
folds that form deep cavities (almost a pocket) where mites
remained well protected. In contrast, U. exsul has shallow folds
on the neck and near the shoulders, which probably do not

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

provide good protection for the chiggers and it make them
susceptible to be detached during the lizards’ sand burrowing
behavior.

In conclusion, mechanisms explaining the disparity in the
occurrence and intensity of chigger infestations on two lizard
species can be complex. Different neck fold morphology,
burrowing behavior and microclimate selection each likely
contributes to the differences observed between the species
considered in this analysis. Within U. stejnegeri, differences in
chigger levels between males and females may be explained by
the much larger home ranges traveled by the males. This species’
lack of sand burrowing behavior, and deeper neck folds and
selection for cooler microhabitats with higher relative humidity
could explain the observed higher chigger infestation levels
compared to U. exsuL Although this study was restricted to
November, infested lizards of both species were observed in
April and August. Seasonal changes in infestation intensity of
chiggers on U. exsul and U. stejnegeri are likely because of
climatic factors (temperature and humidity) influence (Sasa 1961;
Klukowski 2004). Further studies are needed to elucidate these
host-parasite relationships.

Acknowledgements

To Consejo Nacional de Ciencia y Tecnologia (CONACyT)
for the economical support during doctoral programs of CGP and
GC. To Nixon Wilson (University of Northern Iowa) for the
determination of the species of chigger mite, and to A. Rios-
Saldana, G. Mata-Flores, and A. Sanchez-Almazan for their
valuable help in the field.

Literature Cited

Clopton, R. E. & R. E. Gold. 1993. Distribution and seasonal and diurnal activity
patterns of Eutrombicula alfreddugesi (Acari: Trombiculidae) in a forest edge
ecosystem. J. Med. Entomol., 30:47-53.

GARCIA-DE LA PENA, CASTANEDA & BARROWS

Cunha-Barros, M. & C. F. D. Rocha. 2000. Ectoparasitism by chigger mites
{Eutrombicula alfreddugesi: Trombiculidae) in a restinga lizard community.
Ciencia e Cultura, 52:108-1 14.

Cunha-Barros, M., M. Van Sluys, D. Vrcibradic, C. A. B. Galdino, F. H. Hatano
& C. F. D. Rocha. 2003. Patterns of infestation by chigger mites in four
diurnal lizard species from a restinga habitat (Jurubatiba) of Southeastern
Brazil. Braz. J. Biol., 63(3):393-399.

Daniel, M. & A. A. StekoFnikov. 2004. Chiggers mites of the genus
Eutrombicula Ewing, 1938 (Acari: Trombiculidae) from Cuba, with the
description of three new species. Folia Parasit., 51 :359-366.

Davis, J. & R. G. Ford. 1983. Home range in the western fence lizard
{Sceloporus occidentalis occidentalis). Copeia, 1983:933-940.

Garcia, E. 2004. Modificaciones al sistema de clasificacion climatica de Koppen.
Institute de Geografia, Universidad Nacional Autonoma de Mexico. Num. 6,
Mexico, 90 pp.

Garcia-De la Pena, C., G. Castaneda & D. Lazeano. 2005a. Observations on
ectoparasitism by Eutrombicula alfreddugesi (Acari: Trombiculidae) in a
population of Sceloporus cyanogenys. Bull. Chicago Herp. Soc., 40(3):52-53.

Garcia-De la Pena, C., G. Castaneda & D. Lazeano. 2005b. Sceloporus olivaceus
(Texas Spiny Lizard). Ectoparasitism. Herp. Rev., 36(2): 183.

Garcia-De la Pena, C., G. Castaneda, H. Gadsden & A. J. Contreras-Balderas.
2007. Niche segregation within a dune lizard community in Coahuila, Mexico.
Southwest. Nat., 52(2):25 1-257.

INEGI (Instituto Nacional de Estadistica, Geografia e Informatica). 1988. Atlas
Nacional del Medio Fisico. Mexico. 224 pp.

Klukowski, M. 2004. Seasonal changes in abundance of host-seeking chiggers
(Acari: Trombiculidae) and infestations on fence lizards, Sceloporus
undulatus. J. Herpetol., 38(1): 141-144.

Lareschi, M., J. Notamicola, G. Navone & P. M. Linardi. 2003. Arthropod and
filarioid parasites associated with wild rodents in the Northeast Marshes of
Buenos Aires, Argentina. Mem. Inst. Oswaldo Cruz, Rio de Janeiro,
98(5):673-677.

Loomis, R. B. & W. J. Wrenn. 1984. Systematics of the pest chigger genus
Eutrombicula (Acari: Trombiculidae). Pp. 152-159, in Acarology VI, vol. 1.
(D. A. Griffiths & C. E. Bowman, eds.), Wiley, New York, 646 pp.

Potts, J. 2001. Eradication of ectoparasites in children. How to treat infestations
office, scabies and chiggers. Postgrad. Med., 1 10(l):57-64.

Rzedowski, J. 1978. Vegetacion de Mexico. Ed. Limusa. Mexico. 432 pp.

Salvador, A., J. P. Veiga, J. Martin, P. Lopez, M. Abelenda & M. Puerta. 1996.
The cost of producing a sexual signal: testosterone increases the susceptibility
of male lizards to ectoparasite infestation. Behav. EcoL, 7:145-150.

Salvador, A., J. P. Veiga & E. Civantos. 1999. Do skin pockets of lizards reduce
the deleterious effects of ectoparasites? An experimental study with
Psammodromus algirus. Herpetologica, 55(1): 1-7.

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

Sasa, M. 1961. Biology of chiggers. Ann. Rev. Entomol., 6:221-244.

SEMARNAT, Secretana de Medio Ambiente y Recursos Naturales. 2001.
Norma Oficial Mexicana (NOM-059-ECOL-2001). Proteccion ambiental-
Especies nativas de Mexico de flora y fauna silvestres-Categorias de riesgo y
especificaciones para su inclusion, exclusion o cambio-Lista de especies en
riesgo. Diario Oficial de la Federacion (6 de marzo del 2002), Mexico, D.F.

Stebbins, R. C. 1944. Some aspects of the ecology of the iguanid genus Uma.
Ecol. Monogr., 14:311-332.

Talleklint-Eisen, L. & R. J. Eisen. 1999. Abundance of ticks (Acari: Ixodidae)
infesting the western fence lizard, Sceloporus occidentalism in relation to
environmental factors. Exp. Appl. Acarol., 23:731-740.

CG at: cristina.g.delapena@gmail.com

TEXAS J. OF SCI. 59(l):33-38

FEBRUARY, 2007

HELMINTH PARASITE ASSEMBLAGES IN
BULLFROGS {RAN A CATESBEIANA) FROM SOUTHEAST TEXAS

H. Randall Yoder and G. Whitney Gomez

Department of Biology, Lamar University
Beaumont, Texas 77710

Abstract.-A total of 45 bullfrogs {Rana catesbeiana) were collected from Big
Hill Bayou, Jefferson County, Texas in June, 2004 and inspected for helminth
parasites. Forty-two of 45 (prevalence = 93%) were infected with one or more
helminth parasites representing three phyla (three nematode species, six trematode
species, and a single acanthocephalan). A total of 5812 individual helminths were
recovered with mean abundance of infection = 129.16 ± 130.96 worms per host and
mean species richness = 2.22 ± 1.28 helminth species per host individual. This is the
first report of Neoechinorhynchus sp. from Texas bullfrogs.

The bullfrog {Rana catesbeiana) is the largest and one of the
most aquatic ranid frogs occurring in North America. Its range
encompasses a large area from southern Canada to central Florida
in the east through the central U.S. west to eastern Colorado and
eastern New Mexico, and it has been introduced in several other
locations throughout the world (Conant & Collins 1998). Surveys
of endohelminth communities in R. catesbeiana are fairly numer¬
ous. Andrews et al. (1992) provides a North American checklist of
helminth species, locations, and authors for this host. Notable work
omitted or published since that summary includes Harwood (1932),
Muzzall (1991), McAlpine (1997), McAlpine & Burt (1998), and
Goldberg et al. (1998). Relatively few studies concerning these
parasitic associations have been conducted in Texas (Harwood
1932; Knight et al. 1965; Morrison 1966; 1968; Slagle 1966; Hollis
1972). The following adds to this body of information.

Materials and Methods

A total of 45 Bullfrogs {Rana catesbeiana) was collected by
hand from Big Hill Bayou (29°50'N, 95°05'W), Jefferson County,
Texas on June 3, 2004. Frogs were placed on ice and transported to
the laboratory. The frogs, already dead or in a greatly reduced
metabolic state, were placed in a freezer until necropsies were

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

performed. Body surfaee, mouth, eustaehian tubes, body eavity,
lungs, stomach, small intestine, colon, urinary bladder, liver,
kidneys, mesentery, and leg musculature were examined for para¬
sites under a stereomicroscope. Host sex was determined by
observation of reproductive organs. All parasitic helminths were
preserved, stained (when necessary), and mounted using standard
techniques. Identifications of helminth taxa were carried out using
descriptions from the literature (Hedrick 1935; Ward 1940;
Schacher & Crans 1973; Brooks 1976; Kennedy 1981; Prudhoe &
Bray 1982; Baker 1986; Esslinger 1986). Voucher specimens were
deposited in the Harold W. Manter Laboratory, University of
Nebraska State Museum (accession number P-2006-813, HWML
48348-48355). Use of ecological terms follows Bush et al. (1997).

Results

Of the 45 bullfrogs collected, 24 were male, 20 female, and one
was undetermined. Mean snout-vent length was 12.50 ± 1.40 cm
(range = 8.77-16.43 cm). Forty two of 45 (prevalence == 93%) were
infected with one or more helminth parasites. A total of 5812
individual helminths were recovered with mean abundance of
infection = 129.16 ± 130.96 worms per host. Mean species richness
= 2.22 ± 1.28 helminth species per host individual. The component
parasite community of bullfrogs from this location consisted of
three nematode species, six trematode species, and a single
acanthocephalan (Table 1). Larval and adult stages occurred in a
variety of locations throughout the host body (Table 1). The
helminth occurring in the greatest number of hosts (prevalence =
84%) and with highest mean abundance (92.6 ± 92.78 worms per
host) was the larval nematode Spiroxys sp. Because only single
specimens of Falcaustra sp. and Neoechinorhynchus sp. were
found, identification was carried out only to genus. Foleyellides sp.
and the larval Spiroxys sp. were also identified only to genus
because of the inability to distinguish the morphological traits
necessary for species level identification. This was most likely due
to tissue damage resulting from freezing. To the knowledge of the
authors, this is the first report of Neoechinorhynchus sp., or any

YODER & GOMEZ

Table 1. Location of infection, % prevalence, mean abundance ± standard deviation
and range of infection of helminth parasites from Rana catesbeiana («=45).

Parasite taxon Location f % Prevalence Mean abundance Range

in hosts ± standard deviation

Nematoda

Falcaustm sp.

0.02 ±0.15

0-1

Foleyellides sp.

BC/M

0.51 ±0.94

0-4

Spiroxys sp.*

92.6 ± 92.78

0-263

Platyhelminthes

Digenea

Gorgodera amplicava

23.56 ±98.93

0-643

Haematoloechus breviplexus

1.73 ±4.42

0-24

Megalodiscus temperatus

0.04 ± 0.30

0-2

Clinostomum sp.*

0.80 ±5.37

0-36

Metacercaria 1*

9.89 ±21.45

0-107

Metacercaria 2*

1.02 ±6.86

0-46

Acanthocephala

Neoechinorhynchus sp.

0.02 ±0.15

0-1

* Larval stage

t BC = body cavity, C = colon, L=

= lung, M

= mesentery, SI

= small intestine, UB

= urinary

bladder.

other adult acanthocephalan infecting R. catesbeiana from Texas
but McAlpine (1996) reported Neoechinorhynchus rutili adults
from R. catesbeiana in New Brunswick, Canada. Hollis (1972) has
reported Centrorhynchus sp. cystacanth larvae from east Texas
bullfrogs.

Discussion

Parasite communities in bullfrogs from this location were
depauperate and isolationist in nature and largely consistent in
species composition and values of parasitism with others reported
in the literature for Texas and elsewhere. Results from the current
study depart slightly from some studies which report higher rates of
parasitism by terrestrial nematodes (Hollis 1972; McAlpine 1997;
McAlpine & Burt 1998).

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

Few adult helminths eneountered in this study inhabited the
same specific organ within the host. An exception was the use of
the colon by both Falcaustra sp. and Megalodiscus temperatus.
However, these species both occurred at extremely low prevalence
and abundance, rendering the likelihood of interaction very low.
Foleyellides sp. did occur in the body cavity and organ mesentery
along with Spiroxys sp. and the three taxa of metacercaria. The
potential for interaction among these helminths, four of which are
encysted forms, is unknown and worthy of further study.

Species composition of the parasite component community was
consistent with the aquatic nature of the host. The majority of
parasite species had indirect life-cycles with intermediate hosts that
were aquatic during at least part of their life history. With the
possible exception of Falcaustra sp. whose lifecycle is uncertain,
there were no direct life-cycle parasites in the component
community. Recruitment of most helminths occurred via food-web
interactions. This is consistent with the large gape size and
voraciously predatory habits of the bullfrog host which is known to
ingest “nearly anything that moves and that it can swallow”
(Conant & Collins 1998). Exceptions to this generalization include
trematode metacercaria that penetrate the skin of the host and
Foleyellides sp. which infects frogs when mosquitoes feed upon the
blood of the host.

Acknowledgements

The authors would like to thank Lamar University student Pete
O’ Donald for assistance in frog collection and then Hardin-
Jefferson High school student Andy West (currently Lamar
University student) for assisting with necropsies. We also thank
Dr. Richard Harrel for reviewing the manuscript prior to
submission.

Literature Cited

Andrews, K. D., R. L. Lampley, M. A. Gillman, D. T. Corey, S. R. Ballard, M. J.
Blasczyk & W. G. Dyer. 1992. Helminths of Rana catesbeiana in Southern

YODER & GOMEZ

Illinois with a checklist of helminths in bullfrogs of North America. Tran. Illinois
State Acad. Sci., 85(3 and 4):147-172.

Baker, M. R. 1986. Falcaustra species (Nematoda: Kathlaniidae) parasitic in turtles
and frogs in Ontario. Can. J. Zook, 64(1):228"237.

Brooks, D. R. 1976. Parasites of amphibians and reptiles of the great plains. Part 2,
platyhelminths of amphibians in Nebraska. Bull. Univ. Nebr. St. Mus., 10(2):65-
92.

Bush, A. O., K. D. Lafferty, J. M. Lotz & A. W. Shostak. 1997. Parasitology meets
ecology on its own terms: Margolis et al. revisited. J. ParasitoL, 83(4):575-583.

Conant, R. C. & J. T. Collins. 1998. Reptiles and Amphibians: Eastem/Central
North America. Houghton Mifflin Company. New York, 616 pp.

Esslinger, J. H. 1986. Redescription of Foleyellides striatus (Ochoterena and
Caballero, 1932) (Nematoda: Filarioidea) form a Mexican Frog, Rana
montezumae, with reinstatement of the genus Foleyellides Caballero, 1935. Proc.
Helminthol. Soc. Wash., 53(2):2 18-223.

Hedrick, L. R. 1935. The life history and morphology of Spiroxys contortus
(Rudolphi); Nematoda: Spiruridae. Trans. Am. Micr. Soc., 54(4):307-335.

Goldberg, S. R., C. R. Bursey & H. Cheam. 1998. Helminths of two native frog
species {Rana chiricahuensis, Rana yavapaiensis) and one introduced frog
species {Rana catesbeiana) (Ranidae) from Arizona. J. ParasitoL, 84(1): 175-177.

Harwood, P. D. 1932. The helminths parasitic in the Amphibia and Reptilia of
Houston, Texas, and vicinity. Proceedings of the United States National
Museum. 81: 1-74.

Hollis, P. D. 1972. A survey of parasites of the bullfrog, Rana catesbeiana Shaw, in
central east Texas. Southwest. Nat., 17(2): 198-201.

Kennedy, M. J. 1981. A revision of species of the genus Haematoloechus Looss,
1899 (Trematoda: Haematoloehidae) from Canada and the United States. Can. J.
Zook, 59(9):1836-1846.

Knight, M. T., C. J. Barbay & E. O. Morrison. 1965. Incidence of infection by lung-
fluke {Haematoloechus) of the bullfrog, Rana catesbeiana, in Jefferson County,
Texas. Southwest. Nat., 10(2):141-142.

McAlpine, D. F. 1996. Acanthocephala parasitic in North American amphibians: a
review with new records. Alyates, 14(3):1 15-121.

McAlpine, D. F. 1997. Helminth communities in bullfrogs {Rana catesbeiana),
green frogs {Rana clamitans), and leopard frogs {Rana pipiens) from New
Brunswick, Canada. Can. J. Zook, 75(1 1): 1883- 1890.

McAlpine, D. F. & M. D. Burt. 1998. Helminths of bullfrogs, Rana catesbeiana,
green frogs, Rana clamitans, and leopard frogs, Rana pipiens in New Brunswick.
Can. Field-Nat. 1 12(1): 50-68.

Morrison, E. O. 1966. Crayfish, possible secondary intermediate hosts for lung
flukes {Haematoloechus) of bullfrogs in Jefferson County, Texas. Yearbook
Amer. Phil. Soc., 1966:361-362.

Morrison, E. O. 1968. Lungworms of Texas Amphibia. Yearbook Amer. Phil. Soc.,
1968:320-321.

Muzzall, P. M. 1991. Helminth infracommunities of the frogs Rana catesbeiana and
Rana clamitans from Turkey Marsh, Michigan. J. ParasitoL, 77(3):366-371.

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

Prudhoe, S. & R. A. Bray. 1982. Platyhelminth parasites of the Amphibia. Oxford
University Press. Oxford, U.K., 2 1 7pp.

Schacher, J. F. & W. J. Crans. 1973. Foleyella flexicauda sp. n. (Nematoda:
Filarioidea) from Rana catesbeiana in New Jersey, with a review of the genus
and erection of two new subgenera. J. Parasitol., 59(4):685-691.

Slagle, W. G. 1966. A survey of the helminths parasitic in Rana catesbeiana Shaw
and Rana pipiens Schreber from Brazos County, Texas, and vicinity.
Unpublished M.S. thesis, Texas A&M Univ., College Station, 45 pp.

Ward, H. L. 1940. Studies on the life history of Neoechinorhynchus cylindratus
(Van Cleave, 1913) (Acanthocephala). Trans. Am. Micr. Soc., 59(3):327-347.

HRY at: hryoder@my.lamar.edu

TEXAS J. SCI. 59(1 ):39-50

FEBRUARY, 2007

INVENTORY OF THREE BEETLE SPECIES ASSEMBLAGES
(COLEOPTERA: CARABIDAE, SCARABAEOIDEA,
TENEBRIONOIDEA) FROM THE CHIHUAHUAN DESERT
OF WEST TEXAS

Stephanie M. Middleton, Greg H. Broussard*, Michael M. Shipley
and Roy C. Vogtsberger

Department of Biology, Midwestern State University
Wichita Falls, Texas 76308
^Current address

Department of Entomology, Oklahoma State University,

Stillwater, Oklahoma 74078

Abstract.-Pitfall traps were used to study the abundance and diversity of three
species assemblages of Coleoptera from four proximal sites in the Chihuahuan Desert
of the Trans-Pecos region of west Texas. Members of the family Carabidae and
superfamilies Scarabaeoidea and Tenebrionoidea were collected from four ecological
habitats from September 1999 to September 2000. A total of 994 individuals,
representing 53 species of beetles, was collected over the duration of the survey. The
tenebrionoids were the dominant assemblage collected, and were only outnumbered
by the scarabaeoids at one collecting site (Lower Spring). Carabidae was the least
abundant of the three assemblages. The four localities yielded relatively equivalent
collections of individuals, yet Camp Site was the least taxonomically diverse and
exhibited a beetle fauna consisting primarily of tenebrionoids. Collection sites with
water to support a more diverse flora (Sandy Canyon and Upper Spring) also
supported a more diverse beetle fauna. This study provides a baseline for continuing
arthropod investigations of the desert habitats of the Dalquest Research Site in west
Texas.

The order Coleoptera is the largest and most diverse assemblage
of inseets (Rieske & Buss 2001). The sheer numbers and
taxonomie diversity of beetles eommonly dietate that most beetle
studies are either restricted geographically or focus on a limited
number of families (Arnett & Thomas 2001). Only two works are
available that emphasize Chihuahuan Desert beetles of the Trans-
Pecos region: a comprehensive study of beetles from a sand dune
in Big Bend National Park (Dajoz 2000); and a community
assessment of carrion arthropods (Schoenly & Reid 1983).
However, beetles comprise an important part of the Chihuahuan
Desert arthropod fauna, and have proven to be useful environmental

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

indicators of the region, documenting climatic shifts from the late
Pleistocene through Holocene times (Elias & Van Devender 1990).

Pitfall trapping has long been an acceptable method of collecting
terrestrial arthropods (Aheam 1971; Reiske & Buss 2001), in spite
of a confounding array of potential biases (Briggs 1960; Thomas &
Sleeper 1977; Spence & Niemela 1994; Ward et al. 2001). Never¬
theless, this technique provides the opportunity for continuous
sampling where the objectives are to generate faunal inventories
and to permit assemblage comparisons (Faragalla & Adam 1985).

In 1996, the late Walter W. Dalquest and wife Rose donated 518
ha of land to Midwestern State University for the purpose of
fostering biological research. Designated the Dalquest Research
Site (DRS), the property straddles the Brewster/Presidio County
line and extends along the northern boundary of Big Bend Ranch
State Park (Fig. 1). The area is an interface between Bandera Mesa
and a rift valley of heavily eroded canyon lands that shelter isolated
seeps, springs, and intermittent streams. Vegetation is typical
desert scrub, although plant growth is comparatively lush near
water sources.

This study stems from an earlier assessment of cursorial spiders
at the Dalquest Research Site (Broussard & Homer 2006). This
study reports the results of a year of continuous pitfall sampling of
beetle diversity from four ecologically diverse sites, emphasizing
the three most prominent coleopteran assemblages present:
Carabidae (ground beetles and tiger beetles); Scarabaeoidea (June
beetles, dung beetles, and related forms) and Tenebrionoidea
(darkling beetles and allies).

Study Site

Four collection localities were selected on the basis of variation
in substrate, water availability, and resident vegetation. Three sites
were selected from the comparatively heavily vegetated depths of
the canyons, and one site was selected as representing the desert
scmb dominating the mesa and surrounding level terrain.

MIDDLETON ET AL.

Figure 1. Location of Dalquest Research Site (DRS) in Brewster and Presidio counties,
Trans-Pecos Texas. Reference localities included are Big Bend Ranch State Park
(BBRSP), Big Bend National Park (BBNP), and Black Gap Wildlife Management
Area (BGWMA).

The first locality (Camp Site; 29^ 33.408’ N, 103° 47.648’ W;
elevation 1,267 m) is desert hardpan with shallow rocky soils and
exposed bedrock, situated at the fault rim. Terrain is gently rolling
and dominated by creosote (Larrea tridentata (Sesse & Moc. ex
DC.) Coville), lecheguilla (Agave lechuguilla Torr.), and cacti
(Opuntia spp.). Vegetation seldom exceeds 1 m in height. This
area serves as camp headquarters for field investigators and is the
most heavily impacted of all selected sites by human activity,
although disturbance is minimal.

The second site (Sandy Canyon; 29° 33.130’ N, 103° 47.600’ W;
elevation 1,212 m) is situated about 400 m south of Camp Site and
derives its name from the sandy flood plain of an intermittent
stream, the Alamo de Ceserio. Woody vegetation includes
mesquite (Prosopis glandulosa Torr.), little walnut (Juglans
microcarpa Berk), allthom (Koeberlinia spinosa Zucc.), agerito
(Berberis trifoliata Moric.), and oak (Quercus spp.). Clump grasses
and leaf litter contribute to the habitat.

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

Two Other sites less than 900 m to the west in sheltered canyon
bottoms were also selected. One of these (Lower Spring; 29'' 33.391’
N, 103" 47.133’ W; elevation 1,148 m) is sparsely vegetated due to
occasional flooding, with a few scattered creosote shrubs. Soils are of
fine silts and sediments resulting from outflow of the stream. The
fourth site (Upper Spring; 29" 33.165’ N, 103" 47.151’ W; elevation
1,150 m) is approximately 400 m upstream from Lower Spring. Soils
are primarily sands and fine clay that support the most diverse and
dense flora of the study area. The most conspicuous elements are a
mesquite-oak woodland that includes catclaw mimosa {Mimosa
biuncifera Benth.), yucca {Yucca faxoniana (Trel.) Sarg.), desert
sumac {Rhus microphylla Engelm. ex Gray), ocotillo {Fouquieria
splendens Engelm.), graythom lotebush {Ziziphus obtusifolia (Hook,
ex Torr. & Gray) Gray), and beargrass {Nolina erumpens (Torr.)
Wats.).

Materials and Methods

Ten pitfall traps with 10 cm-diameter mouths were placed at each
of the four sites in early September 1999. Each trap consisted of two
one-liter plastic cups, one inside the other, and buried with tops flush
with the ground surface. Each was partially filled with propylene
glycol solution (commercially available low-toxicity antifreeze) to
serve as a killing agent and preservative. Pitfall traps were separated
by approximately 9 m, arranged in grids where terrain permitted, or
placed in linear transects where necessary.

The survey period extended until 15 September 2000, with three
1999 collection dates (7 October, 1 1 November, and 9 December) and
five dates in 2000 (7 January, 24 February, 19 May, 6 July, and 15
September). Collecting effort was recorded as trap nights, which are
calculated as follows: number of pitfall traps times number of
collection sites times number of collecting dates. On each collecting
date, pitfall trap contents were returned to the laboratory at
Midwestern State University for identification. Members of the
Carabidae, Scarabaeoidea (Scarabaeidae, Hybosoridae, Trogidae), and
Tenebrionoidea (Tenebrionidae, Zopheridae) were sorted and
identified to genus based primarily on the keys of Arnett & Thomas
(2001) and Arnett et al. (2002) and, where possible, to species or

MIDDLETON ET AL.

morphospecies by specimen comparison and/or specialist confirma¬
tion. For purposes of discussion, individual species were categorized
as “abundant” {n>15), “common” (/7=30-74), “uncommon” (^=10-29),
and “rare” or of “incidental occurrence” {n=\-9).

Assessment of seasonal availability was accomplished by pooling
collections into three temporal samples: a period roughly equating to
the fall and early winter months (4 September to 9 December, 1999);
an interval encompassing the late winter and spring months (10
December, 1999 to 18 May, 2000), and the duration of the study that
included the summer months (19 May to 14 September, 2000). All
specimens were pinned or point-mounted and are deposited in the
invertebrate collection of the Department of Biology, Midwestern
State University.

Results

A total of 15,070 trap nights collectively produced 994 adult
specimens (Table 1) of the family Carabidae (11 genera, 14 species),
and superfamilies Scarabaeoidea (8 genera and 13 species of
Scarabaeidae; one species of Hybosoridae; 2 genera and four species
of Trogidae) and Tenebrionoidea (14 genera and 20 species of
Tenebrionidae; one species of Zopheridae). The dominant family was
the Tenebrionidae, which comprised nearly half of all beetles taken,
and which was well represented throughout the year at each of the
sites (Table 2). Only at the Lower Spring did another assemblage
(Scarabaeoidea) outnumber the tenebrionoids at any of the four
collection sites during the duration of the study (Table 1). Members
of the Scarabaeoidea comprised 31.3 % of the sample, and were most
commonly taken during the fall period of 4 September-9 December.
Carabidae was the most poorly represented taxon of the three species
assemblages at 18.5 %, with more than half of the specimens collected
from the interval of 19 May to 14 September.

Each of the four collection localities yielded comparable collec¬
tions numerically, although the three canyon collections were more
diverse taxonomically (Fig. 2). Camp Site (^=212) supported 11
genera and species, with three tenebrionoid taxa comprising the
majority of specimens. The majority of tenebrionoids (^=116) were

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

Table 1. Species composition of three species assemblages of Coleoptera collected by pitfall
trapping at Dalquest Research Site, from September 1999 through September 2000.

Taxon

Camp

Site

Sandy

Canyon

Lower

Spring

Upper

Spring

Subtotal

Carabidae

Amara sp

Apenes sp. 1

Aperies sp. 2

Calosoma affine Chaudoir

Calosoma parvicolle Fall

Cymindis sp.

Helluomorphoides sp.

Microlestes sp.

Pasimachus sp.

Pentagonica sp.

Poecilus sp.

Selenophorus sp. 1

Selenophorus sp. 2

Tetragonoderus intersectus
(Germar)

Subtotal

184

Scarabaeoidea

Scarabaeidae

Ataenius convexus Robinson

Ataenius desertus Horn

Ataenius sp.

Canthon imitator Brown

134

146

Diplotaxis brevicornis Cazier

Diplotaxis sp. 1

Diplotaxis sp. 2

Diplotaxis subangulata

LeConte

Euoniticellus intermedins
(Reiche)

Onthophagus velutinus Horn

Oxygrylius ruginasus
(LeConte)

Phyllophaga pusillidens Fall

Serica porcula Casey

Hybosoridae

Hybosorus illigeri Reiche

Trogidae

Omorgus infiatus (Loomis)

Omorgus punctatus (Germar)
or O. infiatus (Loomis)

Omorgus suberosus (Fabricius)

Trox spinulosus Robinson

Subtotal

199

311

MIDDLETON ET AL.

Table 1. Continued.

Taxon

Camp

Site

Sandy

Canyon

Lower

Spring

Upper

Spring

Subtotal

Tenebrionoidea

Tenebrionidae

Argoporis rufipes nitida Casey

Asbolus sp. 1

Asbolus sp. 2

Asidina furcata (Champion)

Asidina paralella (LeConte)

Blapstinus sp. 1

Biapstinus sp. 2

Blapstinus sp, 3

Cryptoglossa infausta (LeConte)
or C texana Blaisdell

Eleodes sp. 1

Eleodes sp. 2

131

Eupsophulus castaneus (Horn)

Eusattus pons Triplehom

Helops sp.

Megasida obliterata (Champion)
or M tenuicollis Triplehorn

Platydema micans Zimmerman

Hylocrinus ^^JSteriphanus sp.

Trimytis sp.

Triorophus sp.

Species undetermined

Zopheridae

Zopherus championi Triplehorn

Subtotal

207

121

499

Total

212

286

292

204

994

taken from 10 December, 1999 to 18 May, 2000. Sandy Canyon
(w=286) supported 23 genera and 33 species, with comparable
representation of the three species assemblages. Lower Spring
{n=292), with 1 8 genera and 20 species, supported mostly scarabaeoid
taxa, although tenebrionoid beetles were well represented.
Scarabaeoids were least common of beetles at Upper Spring (^=204),
with 22 genera and 26 species recorded.

Discussion

Most of the beetles included in this study are essentially terrestrial
“ground” beetles. While many of these taxa are certainly capable of

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

Table 2. Seasonal availability of three species assemblages of Coleoptera collected by pitfall
trapping at Dalquest Research Site from September 1999 through September 2000.

Taxon

4 Sept.-9 Dec.

1 0 Dec. - 1 8 May 1 9 May- 1 4 Sept.

Total Survey Period

(3870 trap nights*)

(6440 trap nights) (4760 trap nights)

(15,070 trap nights)

Camp Site

Carabidae

Scarabaeoidea

Tenebrionoidea

116

207

Subtotal

118

212

Sandy Canyon

Carabidae

Scarabaeoidea

Tenebrionoidea

121

Subtotal

124

286

Lower Spring

Carabidae

Scarabaeoidea

127

167

Tenebrionoidea

Subtotal

156

292

Upper Spring

Carabidae

Scarabaeoidea

Tenebrionoidea

Subtotal

204

Assemblage Totals

Carabidae

110

216

Scarabaeoidea

139

279

Tenebrionoidea

125

219

155

499

TOTALS

298

350

346

994

*Total trap nights among all four collection sites.

flight, few probably exercise this dispersal ability to any great extent.
Life histories of most member species of these beetle assemblages are
imperfectly known, although generalizations of the life histories of
these groups have been advanced (Stehr 1991; Arnett & Thomas
2001; Arnett et al. 2002).

Carabid adults are active, cursorial foragers that opportunistically
scavenge, browse tender vegetation, and hunt for invertebrate prey,
while larvae are typically carnivorous, pursuing a variety of soft-
bodied invertebrates procured from the cover of rocks, rotted logs, and
in leaf litter. Adult scarabaeoid beetles are typically either phyto¬
phagous or saprophagous. The scarabaeiform, or grub-like, larvae of

MIDDLETON ET AL.

(T) Megasida obliterata

or M. tenuicoUis

(S) Omorgus inflatus ^

(T) Argoporis rufipes

(S) Canthon imitator

(T) Trimytis sp.

(T) Blapstinus sp. 1

v®

(S) Ataenius sp.

(S) Euoniticellus intermedius

(T) Cryptoglossa infausta -

or C. texana
(T) Zopherus championi

✓ N

✓ X

oi^

(S) Omorgus suberosus

(S) Trox spinulosus

(T) Asbolus sp. 2
(T) Asidina paralella

(T) Eupsophulus castaneus

(C) Apenes sp. 2
(C) Calosoma affine
(C) Helluomorphoides sp.
(C) Selenophorus sp. 1
(C) Selenophorus sp. 2
(S) Diplotaxis brevicornis
(S) Diplotaxis sp. 1
(S) Diplotaxis sp. 2
(S) Diplotaxis subangulata

(S) Onthophagus velutinus

(T) Blapstinus sp. 3
(T) Eusattus pons
(T) Helops sp.

(T) Platydema micans

(T) Asidina furcata
(T) Species undet.

(Q Poecilus sp.

(S) Ataenius desertus
(S) Oxygrylius ruginasus
(S) Phyllophaga pusillidens

(S) Hybosorus illigeri

(T) Asbolus sp. 1

(S) Serica porcula

(T) Blapstinus sp. 2
(T) Eleodes sp. 1

(S) Omorgus punctatus

or O. inflatus

(T) Eleodes sp. 2

(T) Hylocrinus spJSteriphanus sp.
(T) Triorophus sp.

Figure 2. Cladistical arrangement of three speeies assemblages of Coleoptera collected
by pitfall trapping at four sites on the Dalquest Research Site, from September 1999
through September 2000. Letters in parenthesis indicate members of the three
assemblages: C=Carabidae; S=Scarabaeoidea; and T=Tenebrionoidea.

some taxa feed on plant roots, and both adult and immature stages of
others feed on the dung of mammals. Flight is probably more
common among species of this assemblage. The adult stage of some
species is short-lived. Tenebrionoids are typically long-lived,

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

nocturnal beetles that include a number of desert specialists. Species
commonly are commensal inhabitants of rodent burrows or the nests
of other insects and birds. Adults are primarily fungivores or are
saprophagous and larvae feed on dead and decaying plant matter
ranging from wood to humus.

Most of the beetle taxa identified in this study can be classified as
either rare and resident or as incidental in occurrence and not
necessarily indicative of resident populations at the chosen research
sites. Of the five taxa found at each of the four sites, perhaps only
Eleodes sp. 2 occurred in sufficient numbers to be considered truly
ubiquitous throughout the study area — from the creosote flats to the
canyon depths. Each of the canyon sites supported unique beetle
species (Lower Spring, n=6\ Upper Spring, n=l\ Sandy Canyon,
w=14), possibly reflecting the low vagility of these animals and the
effectiveness of canyon walls as dispersal barriers.

The thin, rocky soils and exposed bedrock of the Camp Site
locality that support little more than creosote and lecheguilla also
support a primarily tenebrionoid beetle fauna comprised of three
species, Eleodes sp. 2 and unconfirmed species of Trimytis and
Triorophus. Dominance by the darkling beetles is attributed largely to
their capability of utilizing the burrows of rodents, especially those of
the locally common Merriam’s kangaroo rat (Dipodomys merriami
Meams). These extensive labyrinths provide shelter, and the bedding,
food stores, and feces probably afford suitable sustenance for both
adults and larvae. Neither of the two species unique to Camp Site,
Asidina furcata (Champion) and an unidentified tenebrionid, were
large components of the survey and are most likely incidental
collections.

Springs, streams, and a higher water table support a comparatively
lush and diverse flora, and correspondingly diverse beetle fauna, in
the canyon localities of Sandy Canyon and Upper Spring. Essentially
serving as woodland refugia, these sites are dominated numerically by
tenebrionids, although the stabilized soils and accumulated humus and
leaf litter contribute to habitats capable of supporting a great

MIDDLETON ET AL.

taxonomic array of carabids and scarabaeoids at all life stages. In
contrast, regular scouring of the Lower Spring site by flood waters
prevents establishment of a more diverse plant community, while
probably also simultaneously purging the resident invertebrate
populations. While the beetle fauna at this locality possessed the least
diversity of the three canyon sites, it is notable that the dung beetle,
Canthon imitator Brown, which was present at each of the canyon
sites, was present at Lower Spring in greater numbers than any other
beetle species at any locality. Most of these records came from the
first sampling period of 4 September-9 December, which was
probably indicative of a single mass emergence. The scarab Ataenius
desertus Horn and tenebrionid Argoporis rufipes nitida Casey
occurred in numbers only at Upper Spring and Lower Spring,
respectively, and we speculate that the fecal waste of larger mammals
(e.g., javelina, Tayassu tajacu (L.); mule deer, Odocoileus hemionus
(Rafmesque)) frequenting the water source is a locally important
nutrition source.

Future Studies

This faunal study provides a baseline inventory for long-term
investigation of the biology of three prominent coleopteran
components of Chihuahuan Desert habitats on the Dalquest Research
Site. Planned efforts include an assessment of less common beetle
taxa, and collections of free-ranging beetle larvae to distinguish
between resident and transient species of carabids, scarabaeoids, and
tenebrionoids. Monitoring of these taxa is ongoing, and the recent
installation of a weather-recording station by Midwestern State
University will help determine species with potential as environmental
indicators.

Acknowledgments

Financial support for the fieldwork was provided by Midwestern
State University, the late Walter W. Dalquest, and wife. Rose. We
thank Ed Riley of Texas A&M University, James Cokendolpher of
Texas Tech University, and Norman Homer of Midwestern State
University for their critical assistance in the identification of
problematic taxa. William Cook of Midwestern State University

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

kindly identified components of the local flora. Frederick Stangl
contributed meaningful insight into presentation of data and the
manuscript.

Literature Cited

Aheam, G. A. 1971. Ecological factors affecting population sampling of desert
tenebrionid beetles. Am. Midi. Nat., 86(2):385-406.

Arnett, R. H. & M. C. Thomas, eds. 2001. American Beetles. Vol. 1. Archostemata,
Myxophaga, Adephaga, Polyphaga: Staphyliniformia. CRC Press, Boca Raton,
Florida, 443 pp.

Arnett, R. H., M. C. Thomas, P. E. Skelley & J. H. Frank, eds. 2002. American Beetles.
Vol. 2. Polyphaga: Scarabaeoidea through Curculionoidea. CRC Press, Boca Raton,
Florida, 861 pp.

Briggs, J. B. 1960. A comparison of pitfall trapping and soil sampling in assessing
populations of two species of ground beetle (Carabidae). East Mailing Res. Stn. Ann.
Rep., 48:108-112.

Broussard, G. H. & N. V. Homer. 2006. Cursorial spiders (Arachnida: Araneae) in the
Chihuahuan Desert of western Texas, USA. Entomol. News, 1 17(3):249-260.

Dajoz, R. 2000. Les Coleopteres d'une du Big Bend National Park (Texas). Nouv. Rev.
Entomol., 17(4):355-363.

Elias, S. A. & T. R. Van Devender. 1990. Fossil Insect Evidence for Late Quaternary
Climatic Change in the Big Bend Region, Chihuahuan Desert, Texas. Quat. Res.,
34:249-261.

Faragalla, A. A. & E. E. Adam. 1985. Pitfall trapping of tenebrionid and carabid beetles
(Coleoptera) in different habitats of the central region of Saudi Arabia. Z. Angew.
Entomol., 99(5):466-471.

Rieske, L. K. & L. J. Buss. 2001. Influence on diversity and abundance of ground- and
litter-dwelling Coleoptera in Appalachian Oak-Hickory forests. Environ. Entomol.,
30(3):484-494.

Schoenly, K. & W. Reid. 1983. Community stmcture of carrion arthropods in the
Chihuahuan Desert. J. Arid Environ., 6(3):253-263.

Spence, J. R. & J. Niemela. 1994. Sampling carabid assemblages with pitfall traps: the
madness and the method. Can. Entomol., 126(3):88 1-894.

Stehr, F. W., ed. 1991. Immature Insects, Vol. 2. Kendall/Hunt Publ. Co., Dubuque,
Iowa, xvi + 975 pp.

Thomas, D. B. & E. L. Sleeper. 1977. The use of pit- fall traps for estimating the
abundance of arthropods with special reference to the Tenebrionidae. Ann. Entomol.
Soc. Am., 70(2):242-248.

Ward, D. F., T. R. New & A. L. Yen. 2001, Effects of pitfall trap spacing on the
abundance, richness and composition of invertebrate catches. J. Insect Conserv.,
5(l):47-53.

SMM at: michael.shipley@mwsu.edu

TEXAS J. SCI. 59(l):51-60

FEBRUARY, 2007

EFFECT OF HUMAN DISTURBANCE ON
THE ABUNDANCE AND SPATIAL DISTRIBUTION OF
THE ATLANTIC GHOST CRAB (OCYPODE QUADRATA)
(FABRICIUS 1798) ON A TEXAS BEACH

Alan D. Maccarone & Patrick L. Mathews

Biology Department, Friends University
Wichita, Kansas 67213

Abstract.-Atlantic Ghost crab (Ocypode quadrata) activity was investigated
along two sections of beach on the Bolivar Peninsula, Texas. One study site was in
an area of high human activity in which significant alterations to the beach has
occurred, and the other site was in a protected bird sanctuary with very low human
activity and little beach alteration. Using line transects as a sampling technique,
measurements of burrow characteristics were conducted. These included the total
number and density of burrows; the diameter and compass orientation of burrow
openings; the distance of burrow openings to the high water line, and the proximity of
neighboring burrows. A sample of captured crabs was measured along with their
burrows to determine the strength of the relationship between the size of a crab and
its burrow opening. Burrow opening size and compass orientation did not differ
between the two study sites, but burrow density was higher in the low-impact area,
and the location of the burrow openings was further from the water on this protected
beach. The ecological ramifications of these findings are discussed.

Ghost crabs (Decapoda: Ocypode) (Fabricius 1798) are found on
sandy beaches throughout the world’s tropical and subtropical
zones (Brown & McLachlan 1990), and act as scavengers and
predators on other littoral macroinvertebrates (Wolcott 1978).
Ghost crabs construct burrows of varying size and depth where they
spend most of their daylight hours (Hill 1981; Chan et al. 2006).
The location and orientation of these burrows in relation to the
beach may be dictated by both natural conditions (Hill 1981; Turra
2005) and human activities (Steiner & Leatherman 1981; Peterson
et al. 2000). The correlation between ghost crab activity and
anthropogenic alterations to beaches has been considered strong
enough to allow a count of burrow openings to be used as a quick
measure of human impacts on such beaches (Barros 2001).

The widely distributed Atlantic Ghost Crab Ocypode quadrata
(Fabricius) ranges from Rhode Island to Brazil, and is the species

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

found throughout the Gulf Coast, including the Bolivar Peninsula
(Amos & Amos 1985). The beachfront property on much of this
peninsula is extensively developed with homes, motels, and com¬
mercial establishments, and much of the beach itself is open to
human and vehicular traffic. Most such human impacts are absent
in the Bolivar Flats Shorebird Sanctuary, located at the western tip
of the peninsula. This sanctuary, which is maintained by the
Houston Audubon Society, consists of salt marshes, sandy beach
areas, and vegetated uplands, all of which are protected from
vehicular traffic and receive only occasional human foot traffic.

Sampling the density, placement, and spatial distribution of O.
quadrata burrows provides a measure of crab population size, as
well as information about the interactions between this species and
its environment. Comparisons of burrows in protected and unpro¬
tected portions of the same beach provide an opportunity to better
understand the interactions between human activity and key fauna
of a sandy beach ecosystem.

Study Area and Methods

The study was conducted in May 2006 at two sites on the
Bolivar Peninsula, east Texas (29° 40’N, 94° 90’ W) (Fig. 1). The
high-impact site, located at Crystal Beach, was characterized by
large human populations observed walking, fishing, swimming, and
sunbathing along the water; by frequent motorized vehicles that
traveled along a 12-m-wide path just above the high-tide line and
parallel to the water, and by overnight campers and other vehicles
parked on the high beach areas just below the sand dunes. The low-
impact site was located 12 km to the west and inside Bolivar Flats
Shorebird Sanctuary (Fig. 1). A large wooden barrier prevents
motorized vehicles from entering the sanctuary, and warning signs
attached to posts alert pedestrians to stay away from the vegetated
areas where seabirds and shorebirds nest. Few people were
observed walking along the beach at the low-impact site during this
study. Other than levels of human activity and disturbance, both

MACCARONE & MATHEWS

Figure 1. (Inset) Map of southeastern Texas and the Gulf of Mexico. (Main map)
Bolivar Peninsula showing the high-impact study site at Crystal Beach (CB) and the
low-impact study site at Bolivar Flats Shorebird Sanctuary (BESS). Sections along
the beach where transects were made are indicated by bars.

study sites were similar with regard to the slope of the beaeh, and
the presenee of wraek lines and vegetated sand dunes.

At eaeh site, the high-tide mark was clearly defined by a wrack
line comprised of Sargassum and other vegetation that had washed
ashore. In the high-impact area, a random location was chosen
along the wrack line for the first transect. The team then moved
westward along the beach, with additional transects run every 50 m.
Each transect began at the upper edge of the wrack line and ended
when the first sand dune vegetation was reached. A total of 15
transects were run in the high-impact area along 750 m of shoreline.
In the low-impact area, the first transect was begun just inside the
wooden barrier. Proceeding west towards the end of the peninsula.

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

only nine transects could be run at this site because beyond the last
transect, the Sargassum wrack zone widened considerably and the
beach narrowed abruptly. Also, patches of the marsh cord grass
Spartina alterniflora began to appear along the beach, which
indicated a different habitat type. Thus, only 450 m of shoreline
was surveyed at this site.

At both sites, each transect was extended out 3 m from either
side of the line. Starting at the wrack line, three observers stood
abreast and walked the entire transect length. Each active burrow
was flagged as the team made its way along the transect line. For
each transect, the following measurements were made: total transect
length, the number of active ghost crab burrows within the 6-m-
wide swath of the transect, and the distance between each active
burrow and the high-water line. For each burrow, burrow diameter
at its widest point was recorded, as was the compass direction of the
burrow opening and the distance to the nearest neighboring active
burrow. An active burrow was defined as one with signs of recent
activity such as piles of damp sand, footprints, or the presence of
food items, or one where a crab had been observed entering.
Shovels were used to dig crabs from randomly-selected burrows.
Crab carapace width was measured with calipers, and the crab was
then released. To minimize the effects of time of day, tide level,
and other abiotic variables, the two sites were studied simultane¬
ously by different teams.

Two-tailed /-tests were used for between-site comparisons of
transect lengths, the number of active burrows per transect, the
distances of burrows above the high-water line, and nearest-
neighbor distances. Pearson correlation was used to measure the
strength of the relationship between burrow width and crab
carapace width, and between burrow width and burrow distance
above high water. Chi-square tests were used to compare both the
linear distributions of burrows between study sites, and the patterns
of compass directions of burrow openings between sites. One-way
ANOVA was used to compare the actual and expected compass

MACCARONE & MATHEWS

directions of burrow openings. Means and one standard deviation
are reported.

Results

The mean length of transects in the low-impact area (47.0 ±3.2
m) and the high-impact area (45.1 ± 3.1 m) did not differ
significantly (^22 ^ 0.99, P > 0.05). A total of 134 active ghost crab
burrows fell within the nine transects run in low-impact area, or a
mean of 14.9 ±3.9 burrows/transect. This was significantly higher
than the 24 burrows (x = 1.6 ± 2.1 burrows/transect) found among
the 15 transects in the high-impact area (^22 = 11.00, P < 0.0001).
When the 6-m-wide swaths were multiplied by the length of each
transect, the total surface area covered by the 15 transects in the
high-impact area was 6345 m^. This surface area resulted in a mean
of 1 burrow/152.0 m The 3653 m covered in the nine transects in
the low-impact area yielded a burrow density three times greater, or
a mean of 1 burrow/47.4 m^. In the low-impact area, neighboring
burrows averaged 1.9 ± 2.2 m apart, significantly closer than the 5.0
± 6.8 m in the high-impact area (656 = 4.07, P < 0.0001). The
pattern of increasing burrow density with greater distance from the
water was reflected in nearest-neighbor distances, which correlated
inversely with burrow distance from the high-water mark (ri56 = -
0.52, P < 0.0001). This relationship was observed at both study
sites.

The mean distance above high water for burrows in the low-
impact area (x = 37.5 ± 10.7 m) was significantly further away
from the water than were burrows in the high-impact area (x = 27.7
± 9.2 m) (/i56 = 4.21, P < 0.0001). In general, few burrows were
located near the water. Burrow density increased in the middle
distances, and then declined closer to the sand dunes (Fig. 2).
However, the linear distributions of burrows along transects
differed significantly between study sites (x^io == 41.89, P < 0.01).
Burrows in the high-impact area were significantly overrepresented
in the middle distances above the high-water mark (21-35 m), but

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

Distance from High Water Line

Figure 2. Distances from the high-water mark for 24 ghost crab burrows at the high-
impact site and 134 burrows at the low-impact site. The pattern of distributions of
burrows between sites is significantly different based on analysis (P < 0.01).

underrepresented near the sand dunes (41-45 m). Mean burrow
diameter in the low-impact area (3.8 ± 1.7 cm) did not differ from
that in the high-impact area (4.3 ± 2.4 cm) {t\s6 = 0.75, P > 0.05).

The compass directions of all burrow openings were grouped
into those facing North (316M5°), East (46^-135°), South (136°-
225°), and West (226°-315°). There was no difference between the
two study sites in the distributions of burrow orientations (x^3 =
7.09, P > 0.05), and so both sites were combined. With all burrows

MACCARONE & MATHEWS

combined, there was a significant departure from an expected equal
distribution of burrow opening directions (x i = 27.07, P < 0.01).
The 60 east-facing burrows were higher than expected and the 15
west-facing burrows were lower than expected. Neither north- nor
south-facing burrows differed from expected (Fig. 3). Mean
burrow diameter did not differ among the four compass directions
(F3, 154 = 0.90, P > 0.05). A total of eight ghost crabs were
excavated from their burrows, some of which went down > 1 m.
Mean burrow diameter and crab carapace diameter correlated
significantly (ry = 0.92, P < 0.001). However, burrow diameter was
not correlated with distance from high water (ri56 0.05, P > 0.05).

Discussion

The physical characteristics of ghost crab burrows, including
size of opening and compass orientation, were similar between the
two study sites, as were crab carapace widths. Such similarities
were not unexpected for a single crab species living on the same
beach, and are probably a response to physical conditions such as
prevailing wind direction (Hill 1981). These results suggest that
parameters unrelated to anthropogenic beach disturbances are
primarily responsible for dictating these characteristics. However,
significant differences were observed between the two beach areas
in crab burrow density and in the spatial distribution of the burrows
in relation to the high-water line. These differences likely reflect
the limitations on burrow site selection and building behavior
forced on O. quadrata as a result of human activities on and
alterations to the beach (Barros 2001; Moss & McPhee 2006).

If ghost crab burrow arrangement in the Bolivar Flats beach area
can be viewed as unrestricted and natural, then the preferred
distance from the high-water line is significantly higher than is
possible in the Crystal Beach area. There, human activities, pets,
buildings, and artificially steep dune edges force burrows to be
constructed nearer the water. In the high-impact area, compaction
of sand and the danger of being crushed by moving vehicles along
the driving path prevent crabs from inhabiting what might be a sub-

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

50 -

I 40

^ 30

■ High-impact
□ Low-impact
0 Total

North East South West
Orientation of Burrow Opening

Figure 3. Compass orientations of 24 burrow openings at the high-impact site and 134
burrow openings at the low-impact site. Based on analysis, there was no
significant difference between study sites in the pattern of burrow compass
orientations (P > 0.05). When burrows in both sites were combined, there was a
significant departure from an expected equal distribution of burrow opening
directions (P < 0.01).

optimal habitat. These threats further restrict and limit the available
beach habitat (Steiner & Leatherman 1981; Turra 2005). Near¬
water burrow sites were equally rare in both locations, possibly due

MACCARONE & MATHEWS

to the danger of predation or a burrow being flooded if it is placed
too close to the surf zone. The obvious result of these limitations is
a much lower total crab density in high-impact zones than in the
more natural, protected beach areas.

The ability to compare nearly-adjacent beach habitats with such
dramatically different levels of human impact provides a rare
opportunity to quantify the ecological ramifications of artificial
disturbance. As is often the case, alteration of the environment by
human activity has decreased the availability of suitable habitat for
a native species. Because O. quadrata is such an important
component of the shoreline food web, acting as both predator of
other invertebrates and as scavenger (Wolcott 1978), the decreased
density and total population size of this species, as well as the
altered placement of burrows, might have cascading effects on
other beach species.

Ideal habitat for this crab species along the Texas coast appears
to include unobstructed beach space > 40 m from the high-water
line. The present study indicates a strong relationship between the
availability of optimal habitat and higher population densities.
Making more of this high beach area available to ghost crabs might
be possible with relatively modest changes in human land-use
patterns along beaches such as this one. In order to reduce
mortality to another species of ghost crab {O. cordimanus),
management interventions were proposed for some Australian
beaches that would prohibit motorized vehicles from certain areas,
or restrict their use between dusk and dawn (Moss & McPhee
2006). Changes such as these may increase the ecological health
and richness of sandy beach ecosystems.

Acknowledgments

We are grateful to the following students who assisted in data
collection for this study: Jon Romain, Alexis Rudko, Jennifer
Jacobs, Hannah Ruling, Philip Neilsen, James Raney, Amber
Bradley, Jodie Hearlson, Lindsay Fugate, and April Harrison. We

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

also thank two peer reviewers, whose valuable suggestions
improved the manuscript.

Literature Cited

Amos, W. H. & S. H. Amos. 1985. Audubon Society Guide to the Atlantic & Gulf
Coasts. Alfred A. Knopf , New York, NY, 670pp.

Barros, F. 2001. Ghost crabs as a tool for rapid assessment of human impacts on
exposed sandy beaches. Biol. Conserv., 97(3):399-404.

Brown, A. C. & A. McLachlan. 1990. Ecology of sandy shores. Elsevier,
Amsterdam,, 340 pp.

Chan, B. K. K., K. K. Y. Chan & P. C. M. Leung. 2006. Burrow architecture of the
ghost crab Ocypode ceratopthalma on a sandy shore in Hong Kong.
Hydrobiologia, 560:43-49.

Hill, G. W. 1981. Orientation of ghost crab Ocypode quadrata (Fabricius) burrows
as an indicator of shoreline position and wind direction. Texas J. Sci., 34(1):23-
34.

Moss, D. & D. P. McPhee. 2006. The impacts of recreational four-wheel driving on
the abundance of the Ghost Crab {Ocypode cordimanus) on a subtropical sandy
beach in SE Queensland. Coastal Manag., 34:133-140.

Peterson, C. H., D. H. M. Hickerson & G. G. Johnson. 2000. Short-term
consequences of nourishment and bulldozing on the dominant large invertebrates
of a sandy beach. J. Coast. Res., 16(2):368-378.

Steiner, A. J. & S. P. Leatherman. 1981. Recreational impacts on the distribution of
ghost crabs Ocypode quadrata Fab. Biol. Conserv., 20(2): 1 1 1-122.

Turra, A. 2005. Spatial distribution of the ghost crab Ocypode quadrata in low-
energy tide-dominated sandy beaches. J. Nat. Hist., 39(23):2 163-2 177.

Wolcott, T. G. 1978. Ecological role of ghost crabs, Ocypode quadrata (Fabricius),
on an ocean beach: scavengers or predators? J. Exp. Mar. Biol. EcoL, 31(1):67-
82.

ADM at: alanm@friends.edu

TEXAS J. SCI. 59(l):61-72

FEBRUARY, 2007

GPS-BASED ANALYSIS OF SHORELINE CHANGE, 1995-2005, MAD
ISLAND MARSH PRESERVE, MATAGORDA COUNTY, TEXAS.

Webster Mangham and Harry Williams

Trinity River Authority of Texas, 5300 South Collins
Arlington, Texas 76004-0060 and
Geography Department, University of North Texas
Denton, Texas 76203

Abstract.-A combination of a global positioning system (GPS), high-
resolution aerial photographs and a geographic information system (GIS) was
used to measure shoreline change at Mad Island Marsh Preserve during the period
1995 - 2005. The magnitude and direction of shoreline change was found to vary
between five contrasting shoreline types ineluded in the study. Compared to
earlier studies that suggested erosion occurred along all sections of shoreline in
the study area between 1941 and 1993, this study found sections of shoreline that
had essentially stabilized and others where previously documented erosion had
slowed eonsiderably. The reason(s) for these changes is uncertain - two
possibilities are that a progressively-widening wave-cut platform is dissipating
barge-generated wave energy or that the exposure of dense root mats along some
receding shorelines is increasing resistance to wave erosion. A eonerete erosion
eontrol mat emplaced along one section of shoreline in 1996 was found to have
been unsuceessful, because soil had been washed out from under the concrete.

The Texas Nature Conservancy’s Mad Island Marsh Preserve is
a 29-km^ tract containing the east arm of Mad Island Lake, its
associated freshwater and brackish marshes, and surrounding
upland prairie and shrub land habitats. The marshes on the preserve
provide an important habitat for many aquatic organisms and an
important wintering ground for millions of migratory birds that use
the Central Flyway each year. Since 1993, the preserve has ranked
among the top-five areas in the nation in number of species counted
during the annual Audubon Society’s Christmas Bird Count (L.
Halsted, Texas Nature Conservancy, Pers. Comm., 2003). The
shoreline of the preserve borders the Gulf Intracoastal Waterway
(GIWW), a man-made canal constructed in 1941 (Figure la).

Barge traffic in the GIWW has caused considerable wave
erosion of preserve shorelines over the last 65 years. Williams
(1993a) calculated erosion rates along the shores of the preserve

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

based on sequential aerial photographs covering the period 1943-
1991. In 1993 Williams conducted a second study of shoreline
erosion based on repeated measurements of the separation between
the shoreline and a series of survey stakes set in the ground every
30.48 m along the shoreline. The surveys were conducted in June
of 1992 and in May of 1993, and were used to derive 1-year erosion
rates (Williams 1993b). The results of those studies showed long¬
term shoreline erosion approached 2 m/year on some parts of the
preserve. The studies also suggested that erosion may threaten
freshwater marsh habitats bordering Mad Island Lake, since erosive
shortening of Mad Island Bayou - a tidal inlet connecting the lake
to Matagorda Bay - is likely to increase salt water intrusion into the
lake. In response to these findings, the Texas Nature Conservancy
in 1996 constructed a protective concrete barrier along the shoreline
bordering Mad Island Bayou in the hopes of preventing further
erosion (Fig. lb: section C).

The focus of this study is to determine recent shoreline change
along a ca. 1.2 km stretch of shoreline bordering the mouth of Mad
Island Bayou (Figure lb). The objectives of the study are to: (1)
measure shoreline change between 1995 and 2005, using a com¬
bination of high-resolution aerial photographs, GPS and GIS, (2)
compare shoreline change derived for this study with the results of
Williams (1993a; 1993b), (3) evaluate the effectiveness of the
concrete mat emplaced in 1996 in preventing erosion (shoreline C
in Figure lb), and (4) investigate the variation in shoreline change
between the different shoreline types found along this part of the
preserve; these include an area of Spartina marsh growing out into
the GIWW, an area of marsh bordered by a ca. 0.5 m cliff, the
concrete erosion control mat, an area of dense woodland bordered
by a ca. 1.5 m cliff and an area fronted by a gently sloping sandy
beach (Figure lb: sections A to E).

Materials and Methods

1995 shoreline map -A map of the shoreline in 1995 was
derived from a 1 -meter-resolution Digital Ortho Quarter Quad

MANGHAM & WILLIAMS

Fig. 1. (a) Location of Mad Island Island Marsh Preserve, Matagorda County, Texas, (b)
Study area showing Mad Island Bayou and the Gulf Intracoastal Waterway.
Shoreline divisions: A - Spartina marsh; B - Cliffed marsh; C - Concrete mat; D -
Cliffed wooelands; E - Sandy beach. Numbers indicate location of survey stakes
used by Williams (1993b).

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

(DOQQ) downloaded from the Texas Natural Resources
Information System web site (TNRIS 2005). For the purposes of
the study, the shoreline was defined as the edge of the erosional
scarp that borders the majority of the GIWW in the study area.
Where the erosional scarp was not present, the vegetation line was
used. The shoreline was defined in this manner because it is readily
identifiable on air photographs and is consistent with Williams’
(1993a) methods. To improve accuracy and provide assessment of
digitizing error, the shoreline was digitized three times at a scale of
1:500 (large visible pixels), 1:1,000 (medium visible pixels) and
1:1,500 (small visible pixels). The three resulting digitized
shorelines were averaged within the GIS to create a single 1995
shoreline. The principles of the Digital Shoreline Analysis System
(DSAS) extension, written for ArcView™ (Theiler et al. 2003),
were used to create a single shoreline. DSAS is based on creating
regularly spaced perpendicular transects to be used as measurement
locations across multiple shorelines (Theiler et al. 2003). Follow¬
ing this methodology, five-meter-interval transects were created
across the three shorelines and the distance across the shorelines
was measured at each transect. The average distance across the
three shorelines was 0.97 m. Based on this result, the digitizing
error was assumed to be ± 1 m. The GIS was used to create a single
1995 shoreline passing through the average position of the three
digitized shorelines along each transect (Figure 2).

2005 shoreline map -In February of 2005, a sub-meter accuracy
backpack GPS was used to map the shoreline. To improve
accuracy and provide assessment of GPS mapping error, the
shoreline was mapped three times. The GPS files were differ¬
entially corrected and imported into the GIS. The separation of the
three resulting shorelines was measured along transects at 5-m
intervals. The average distance across the three shorelines was 0.69
m. Based on this result, the GPS mapping error was also assumed
to be ± 1 m. The three shorelines were averaged within the GIS to
create a single 2005 shoreline.

MANGHAM & WILLIAMS

^ 94

j I

Fig. 2. Illustration of three shorelines digitized from the 1995 DOQQ, 5-m- interval
transects and the average 1995 shoreline. Distance across the three shorelines is
measured at each transect for error assessment. Transects were also used to help
locate the average shoreline position for creation of a single 1995 shoreline.

Additional error assessment mapping error plays a
critical role in determining the magnitude of shoreline change, two
additional error assessment techniques were employed: the
accuracy of the 1995 DOQQ was further assessed by collecting
GPS positions of a number of Ground Control Points (GCPs) within
the study area. Eleven GCPs, readily loeated in the field and
identifiable as a single pixel on the DOQQ, were selected
(examples include the comer of a building and the center of a road
intersection). The GPS positions were imported into the GIS and
compared to the position of the GCPs on the DOQQ: all 1 1 GPS
points were located within the correct pixel on the map, suggesting
a level of accuracy of ± 1 m.

To further assess the accuracy of GPS-based distance
measurements, a GPS point was recorded for seven survey stakes
remaining from Williams’ 1993b study (stake numbers 1, 3, 8, 9,
10, 11 and 13 - all within section A of the study area (Fig. lb);

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

Table 1 . Comparison of measurements made by tape and GPS.

Measurement

Tape (m)

GPS (m)

Difference (m)

Stake 1 to shoreline

20.0

20.1

0.1

Stake 3 to shoreline

14.9

15.1

0.2

Stake 8 to shoreline

22.1

0.0

Stake 9 to shoreline

21.6

21.7

0.1

Stake 1 0 to shoreline

19.9

19.8

0.1

Stake 1 1 to shoreline

18.2

18.3

0.1

Stake 13 to shoreline

5.4

5.3

0.1

Stake 1 to stake 3

62.0

62.2

0.2

Stake 8 to stake 9

30.4

30.5

0.1

Stake 9 to stake 1 0

30.7

0.0

Stake 1 0 to stake 1 1

30.4

30.2

0.2

Stake 1 1 to stake 1 3

60.9

60.5

0.4

Mean

0.13

Other stakes from Williams’ 1993b study could not be found within
the study area and are presumably missing). The perpendicular
distance from each stake to the shoreline was measured with a 30m
tape and the shoreline position was recorded with a GPS point. The
tape was also used to measure the distances between stakes 1 and 3,
8 and 9, 9 and 10, 10 and 1 1 and 1 1 and 13. The GPS points were
imported into the GIS and distances between points were measured
for comparison to the same distances measured by tape (Table 1).

All the GPS-based measurements were within 0.4 m of the tape
measurements (mean difference was 0.13 m). This result is consis¬
tent with the reported sub-meter accuracy of the GPS and is within
the ± 1 m mapping error derived from shoreline mapping. Overall,
the assessment of error suggests a point accuracy of ± 1 m for both
the DOQQ and GPS-based shoreline maps. This provides a
distance measurement accuracy of ± 2m (each end of a measured
line is considered a point with ± 1 m accuracy).

Shoreline change: 1995-2005 -T\\q GIS was used to create
shoreline-perpendicular transects at 10-m spacing along the
shoreline of the entire study area. The separation of the 1995 and
2005 shorelines was measured along the resulting 110 transect
lines. Although shoreline retreat (erosion) was apparent along

MANGHAM & WILLIAMS

many transects, shoreline accretion (growth of the shoreline out into
the GIWW) was found at approximately one third of measurement
locations.

Discussion

Because one of the objectives of the study is to compare
shoreline change between different shoreline types, the results are
discussed in the context of the shoreline divisions shown in Figure
lb. As the measurement error for distance is ± 2 m, the error for
calculated rates of shoreline change over the 10 year period of the
study is ± 0.2 m/year.

Section ^.-This section of shoreline consists of Spartina marsh
growing out into the GIWW with no visible scarp at the water’s
edge. Changes along this ca. 430-m-long section are a mixture of
erosion and accretion (Figure 3). Along many transects, the
shoreline position between 1995 and 2005 is essentially unchanged,
the erosion or accretion amount falling within the 2-m measurement
error. There are however pockets of erosion and accretion where
shoreline changes exceed the measurement error. The maximum
erosion is between 5.2 and 9.2 m (7.2 ± 2 m). The maximum
accretion is between 2 and 6 m (4 ± 2 m). The average change for
the entire section is erosion between 0.2 and 4.2 m (2.2 ± 2 m;
equivalent to 0.02 to 0.42 m/year); however, the use of an average
is misleading in this instance, because the small pockets of erosion
and accretion, which are clearly present, cancel out when averaged.

These findings contrast with those of Williams (1993a), who
found no evidence of accretion between 1943 and 1991, and
calculated an average erosion rate for this section of shoreline of
1.18 m/year (based on two measurement locations, about 200 m
apart). In a follow-up study, Williams (1993b) found one-year
erosion rates averaging 0.69 m/year for this section of shoreline,
based on 14 measurement locations spaced about 30 m apart.
Williams (1993b) also reported that a small erosive scarp was
present along most of this shoreline in 1993; this scarp was not
observed in 2005.

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

SECTION A

- 2005

- 1995

SECTION B

100
j i j

meters

Fig. 3. Shoreline changes between 1995 and 2005, shoreline sections A - E (shoreline
section locations shown in Fig. lb).

MANGHAM & WILLIAMS

The reason(s) for these changes are unknown; this section of
shoreline appears to be stabilizing: on average, erosion has slowed
considerably and no longer affects the entire section of shoreline.
Shoreline accretion suggests that sediment trapping is occurring in
places, presumably aided by the Spartina marsh. A possible
explanation for diminishing wave erosion is that wave energy is
being dissipated over a progressively widening wave-cut platform -
a possibility suggested by Williams (1993a).

Section -This section of shoreline consists of an area of marsh
bordered by a prominent ca. 0.5-m-high cliff. With the exception
of a small pocket of accretion at the western end, this shoreline has
undergone considerable erosion between 1995 and 2005 (Figure 3).
Maximum erosion is between 9.6 and 13.6 m (11.6 ± 2 m). The
average change for the entire section is erosion between 4.9 and 8.9
m (6.9 ± 2 m; equivalent to 0.49 to 0.89 m/year). These findings
are more in line with those of Williams (1993a; 1993b). Williams
(1993a) calculated an average erosion rate of 1.04 m/year based on
two measurement locations about 200 m apart. Williams (1993b)
calculated an average one-year erosion rate of 1.26 m/year for this
section of shoreline, based on five measurement locations spaced
about 30 m apart.

Section C,-This section of shoreline consists of a thin peninsula
of land bordered by the concrete erosion control mat emplaced in
1996. Shoreline position along this section of shoreline is
essentially unchanged between 1995 and 2005 (Figure 3). The
separation between the 1995 and 2005 shorelines along all 13
transects located in this section is less than the 2 m measurement
error. This is not a surprising result because the concrete mat forms
the shoreline along this section and it has been in place nine out of
the ten years of the study period. Williams (1993a) calculated an
erosion rate of 0.9 m/year for this section of shoreline, based on one
measurement location. Williams (1993b) calculated an average
one-year erosion rate of 0.82 m/year, based on four measurement
locations spaced about 30 m apart. Although it appears that this

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

section of shoreline has stabilized and further erosion has been
prevented, this result is misleading because field observations
indicate that in several places soil underlying the concrete mat has
been washed out, leaving the structure hollow and prone to collapse
(Figure 4).

Section D.-This section of shoreline consists of dense woodland
bordered by a ca. 1.5 m cliff. Shoreline changes along this section
are a mixture of erosion and accretion (Figure 3). Shoreline change
is negligible along most transects, the measured erosion or
accretion amount falling within the 2-m measurement error. The
maximum erosion is between 2.5 and 6.5 m (4.5 ± 2 m). The
maximum accretion is between 4.4 and 8.4 m (6.4 ± 2 m).

Williams (1993a) calculated an erosion rate for this section of
shoreline of 0.61 m/year (based on one measurement location).
Williams (1993b) found one-year erosion rates averaging 0.45
m/year, based on eight measurement locations spaced 15-30 m
apart. These results suggest that erosion has slowed considerably
on this section of shoreline and it has become relatively stable. The
reason(s) for this change is unknown. A dense root mat was
observed in the cliff face bordering this shoreline in 2005 - it may
be that the root mat is helping to strengthen and stabilize the cliff
and retard the effects of wave erosion. The effects of a widening
wave-cut platform may also contribute to a reduction of erosion at
this site.

Section £'.-This 130-m-long section of shoreline consists of a
gently sloping sandy beach. Approximately the western two-thirds
of this shoreline underwent erosion between 1995 and 2005 (Figure
3). The maximum erosion was measured near the center of this
section, where the shoreline retreated between 5.4 and 9.4 m (7.4 +
2 m). Measurements along three transects within the eastern one-
third of this shoreline indicate a small amount of accretion, but the
amount of accretion is either below or close to the 2-m
measurement error. The average change for the entire section is

MANGHAM & WILLIAMS

Fig. 4. Collapsed section of the concrete erosion control mat in Section C. Soil has been
washed out from beneath this section of the mat and tidal waters from the GIWW (in
the foreground) flow freely into Mad Island Bayou (in the background).

erosion between 1.3 and 5.3 m (3.3 ± 2 m; equivalent to 0.13 to
0.53 m/year). These findings suggest erosion has slowed along this
shoreline in the last deeade. Williams (1993a) ealculated an
average erosion rate of 1.06 m/year based on one measurement
location. Williams (1993b) calculated an average one-year erosion
rate of 0.88 m/year for this section of shoreline, based on seven
measurement locations spaced 15-30 m apart. The reason(s) for the
apparent slowing of erosion is unknown - it may also be the result
of a widening wave-cut platform, has as been suggested for other
shoreline sections where erosion appeared to have declined over the
last decade.

Conclusions

Combining a sub-meter-accuracy Global Positioning System and
a 1 -m-resolution DOQQ to map shoreline change provided a level
of point accuracy of ± 1 m and distance accuracy of ± 2 m. The
results of the study suggest that shoreline change varies between

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

different shoreline types and that rates of shoreline ehange over the
last deeade vary from those found by earlier studies. Overall,
erosion appears to have deelined within the study area, with erosion
slowing in some plaees and shorelines beeoming relatively stable in
other plaees. Shoreline seetions A and D, which displayed
consistent erosion between 1943 and 1993, underwent relatively
little change between 1995 and 2005 and appear to have stabilized.
The reason(s) for this change is uncertain - it may be that a
progressively widening wave-cut platform is dissipating wave
energy along section A; whereas a dense woodland root mat may be
retarding wave erosion along the shoreline bordering section D.
Other shoreline sections where erosion has apparently slowed may
also be showing the effects of a widening wave-cut platform. The
shoreline bordered by the concrete erosion control mat showed
negligible change in position between 1995 and 2005. Although
the concrete mat has apparently stabilized this shoreline, this result
is misleading because soil beneath the mat has been washed away
in several places and tidal waters from the GIWW flow freely into
Mad Island Bayou through the lattice structure of the mat. This will
presumably increase saltwater intrusion into Mad Island Lake - an
outcome the mat was designed to prevent.

Literature Cited

Halsted, L. 2003. Texas Nature Conservaney Internal Document: Conservation Plan
for Mad Island Marsh — Oyster Lake, from
conservationonline.org/2003/07/LCAP_MadIsland.

Texas Natural Resource Information System (TNRIS). 1995. DOQQ Map: Palacios
NW.

Theiler, R., D. Martin & A. Ergul. 2003. Digital Shoreline Analysis System (DSAS)
version 2.0: An ArcView extension for calculating shoreline change. United
States Geological Survey.

Williams, H. 1993a. Shoreline Erosion at Mad Island Marsh Preserve, Matagorda
County, Texas. Texas J. Science, 45(4):300-309.

Williams, H. 1993b. Shoreline Erosion Monitoring Network, Mad Island Marsh
Preserve, Matagorda County, Texas: First Year Results (1992 - 1993). The
Nature Conservancy of Texas, Research Report, 12 pp.

HW at: williams@unt.edu

TEXAS J. SCI. 59(1), FEBRUARY, 2007

GENERAL NOTES

DEMISE OF AN INTRODUCED POPULATION OF
THE SAILFIN MOLLY, POECILIA LATIPINNA, IN THE
UPPER GUADALUPE RIVER OF CENTRAL TEXAS

Fred B. Stevens, Adriana S. Jarussi and James Athey

Biology Department, Sehreiner University, Kerrville, TX 78028

Populations of the sailfin molly, Poecilia latipinna, are native to
Mexico and the Rio Grande valley and coastal Texas. Other Texas
populations are considered to be the result of introductions beyond
the natural range for this species (Brown 1953; Hubbs et. al. 1991).

Poecilia latipinna was collected in the upper Guadalupe River
drainage, in Johnson Creek, in Kerr County (30°9’N, 99°20’W) at
the Beach Road crossing (later renamed Byas Springs Road) in
January 1989. Sailfin mollies continued to appear in collections for
several years at the same location. By 1994, the population had
spread into the Guadalupe River where it was collected as far
downstream as Center Point, Texas (29°57’N, 99°2’W; Kerr
County). Mollies were easily observable along this section of the
Guadalupe River as the abundance of mollies also had increased
substantially. At this point, it appeared that P. latipinna had
become a permanent part of the fish fauna of the upper Guadalupe
River.

Subsequent collection efforts in December 1996 did not yield
mollies at either the Johnson Creek or the Center Point locations.
Later collections through September 2006 (Table 1) failed to yield
specimens.

The apparent extinction of an apparently thriving population is
problematic. Cold weather detrimental to this subtropical species is
not involved. The population disappeared between April and
December 1996, and no intervening cold spell occurred. However,

THE TEXAS JOURNAL OF SCIENCE - VOL. 59, NO 1, 2007

Table 1. Collection results for Poecilia latipinna from Johnson Creek, Texas by date.
When higher numbers are present (1992 and April 1996), P. latipinna was also
present in much of the adjacent Guadalupe River, as far downstream as Center Point.

Date

Number of P. latipinna

Jan 20, 1986

Jan 14, 1988

Jan 11, 1989

April 21, 1989

May 5, 1989

May 27, 1989

July 22, 1989

Sept. 28, 1989

Jan. 8, 1990

Jan. 13, 1992

April 10, 1996

December 3, 1996

April 15, 1997

Sept. 15, 2006

there was a significant flooding event during this period and
flooding has been known to exterminate isolated populations of
other livebearer species (Collins et. al. 1981). The U.S. Geological
Survey monitoring station at Kerrville, just downstream from the
area inhabited by P. latipinna, recorded a peak flow of over 55,000
cubic feet per second (cfs) on October 28, 1996 (Fig. 1) (U.S.
Geological Survey, http://nwis.waterdata.usgs.gov/tx/nwis). This
was the most significant high water event since 1987 and 1988.
The mean flow for October based on 20 years of records is 1 63 cfs,
and the flow on October 26, 1996 was 47 cfs. It is presumed that
this dramatic flood event and its associated physical stresses lead to
the demise of the introduced P. latipinna population.

TEXAS J. SCI. 59(1), FEBRUARY, 2007

Li.

40000

30000

20000

10000

Figure 1: Flow of Guadalupe River at
Kerrville From June 1986 to January
1997(USGS Data)

^ ^ j-

Date

- Average Daily

Flow

Acknowledgement

The senior author wishes to thank Texas Parks and Wildlife for
the continuation of scientific permit number SPR-0690-1 17.

Literature Cited

Brown, W. H. 1953, Introduced fish species of the Guadalupe River Basin. Texas J.
Sci., 5(3):245-251.

Collins, J. P., C. Young, J. Howell & W. L. Minckley. 1981. Impact of flooding in a
Sonoran Desert stream, including elimination of an endangered fish population
{Poeciliopsis o. occidentalism Poeciliidae). Southwest. Nat., 26(4):41 5-4223.
Hubbs, C., R. J. Edwards & G. P. Garrett. 1991. An annotated checklist of the
freshwater fishes of Texas, with keys to identification of species. Texas J. Sci.,
Suppl.,43(4):l-56.

FBS at: fstevens(gschreiner.edu

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

THE TEXAS ACADEMY OF SCIENCE

www.texasacademyofscience.org

Membership Information and Application

MEMBERSHIP.-Any person or member of any group engaged in scientific
work or interested in the promotion of science is eligible for membership in The
Texas Academy of Science.

(Please print or type)

Name

Last

First

Middle

Mailing

Address

City_

State

Zip

FAX

E-mail:

Regular Member $30.00

Supporting Member $60.00
Sustaining Member $100.00
Patron Member $150.00

Student-Undergraduate $15.00_
Student-Graduate $ 1 5 .00
Joint $35.00

Emeritus $10.00
Corporate Member $150.00_
Affiliate $5.00
(list name of organization)

Contribution

AMOUNT REMITTED

SECTIONAL INTEREST AREAS:

Anthropology

Botany

Cell & Moleseular Biology
Chemistry & Biochemistry
Computer Science
Conservation Ecology
Environmental Science

Freshwater & Marine Sciences

Geosciences

Mathematics

Physics

Science Education

Systematics & Evolutionary Biology
Terrestrial Ecology & Management

Please indicate your Sectional interest(s) below:

Send Application Form and Check or Money Order to:

Dr. Fred Stevens, Executive Secretary
The Texas Academy of Science
CMB 6252
Schreiner University
Kerrville, Texas 78028-5697

Please photocopy this Application Form

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

INSTRUCTIONS TO AUTHORS

Scholarly manuscripts reporting original research results in any field of
science or technology will be considered for publication in The Texas
Journal of Science. Prior to acceptance, each manuscript will be reviewed
both by knowledgeable peers and by the editorial staff. Authors are en¬
couraged to suggest the names and addresses of two potential reviewers to
the Manuscript Editor at the time of submission of their manuscript. No
manuscript submitted to the Journal is to have been published or submitted
elsewhere. Excess authorship is discouraged. Manuscripts listing more than
four authors will be returned to the corresponding author.

Upon completion of the peer review process, the corresponding author is
required to submit two letter quality hard copies of the final revised manu¬
script as well as an electronic copy.

Format

Except for the corresponding author’s address, manuscripts must be
double-spaced throughout (including legends and literature cited) and
submitted in TRIPEICATE (typed or photocopied) on 8.5 by 11 inch bond
paper, with margins of approximately one inch and pages numbered.
Scientific names of species should be placed in italics. Words should not be
hyphenated. The text can be subdivided into sections as deemed appropriate
by the author(s). Possible examples are: Abstract; Methods and Materials;
Results; Discussion; Summary or Conclusions; Acknowledgments; Eitera-
ture Cited. Major internal headings are centered and capitalized. Computer
generated manuscripts must be reproduced as letter quality or laser prints.

References

References must be cited in the text by author and date in chronological
(not alphabetical) order; Jones (1971); Jones (1971; 1975); (Jones 1971);
(Jones 1971; 1975); (Jones 1971; Smith 1973; Davis 1975); Jones (1971);
Smith (1973); Davis (1975); Smith & Davis (1985); (Smith & Davis 1985).
Reference format for more than two authors is Jones et al. (1976) or (Jones
et al. 1976). Citations to publications by the same author(s) in the same year
should be designated alphabetically (1979a; 1979b).

Eiterature Cited

Journal abbreviations in the Eiterature Cited section should follow those
listed in BIOSIS Previews ® Database (ISSN: 1044-4297). All libraries

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

receiving Biological Abstracts have this text; it is available from the
interlibrary loan officer or head librarian. Otherwise standard recognized
abbreviations in the field of study should be used. All citations in the text
must be included in the Literature Cited section and vice versa.

Consecutively-paged journal volumes and other serials should be cited
by volume, number and pagination. Serials with more than one number and
that are not consecutively paged should be cited by number as well. The
following are examples of a variety of citations:

Journals & Serials -

Jones, T. L. 1971. Vegetational patterns in the Guadalupe Mountains,
Texas. Am. J. Bot., 76(3):266-278.

Smith, J. D. 1973. Geographic variation in the Seminole bat, Lasiurus
seminolus. J. Mammal., 54(l):25-38.

Smith, J. D. & G. L. Davis. 1985. Bats of the Yucatan Peninsula. Occas.
Pap. Mus., Texas Tech Univ., 97:1-36.

Books -

Jones, T. L. 1975. An introduction to the study of plants. John Wiley &
Sons, New York, 386 pp.

Jones, T. L., A. L. Bain & E. C. Bums. 1976. Grasses of Texas. Pp. 205-
265, in Native grasses of North America (R. R. Dunn, ed.), Univ. Texas
Studies, 205:630 pp.

Unpublished -

Davis, G. L. 1975. The mammals of the Mexican state of Yucatan.
Unpublished Ph.D. dissertation, Texas Tech Univ., Lubbock, 396 pp.

In the text of the manuscript, the above unpublished reference should be
cited as Davis (1975) or (Davis 1975). Unpublished material that cannot be
obtained nor reviewed by other investigators (such as unpublished or
unpublished field notes) should not be cited.

Abstraets-

For a professional meeting abstract, give the author(s), year, title and
the website or URL where the abstract is available.

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

Graphics, Figures & Tables

Every table must be included as a computer generated addendum or
appendix of the manuscript. Computer generated figures and graphics must
be laser quality and camera ready, reduced to 5.5 in. (14 cm) in width and no
more than 8.5 in. (20.5 cm) in height. Shading is unacceptable. Instead,
different and contrasting styles of crosshatching, grids, line tints, dot size, or
other suitable matrix can denote differences in graphics or figures. Figures,
maps and graphs should be reduced to the above graphic measurements by a
photographic method. A high contrast black and white process known as a
PMT or Camera Copy Print is recommended. Authors unable to provide re¬
duced PMT*s should submit their originals. Do not affix originals or PMT's
to a cardboard backing. Figures and graphs which are too wide to be
reduced to the above measurements may be positioned sideways. They
should then be reduced to 9 in. (23 cm) wide and 5 in. (12.5 cm) in height.
Black and white photographs of specimens, study sites, etc. should comply
with the above dimensions for figures. Color photographs cannot be
processed at this time. Each figure should be marked on the back with the
name of the author(s) and figure number and top of figure. All legends for
figures and tables must be typed (double-spaced) on a sheet(s) of paper
separate from the text. All figures must be referred to in text as ’’Figure 3”
or ”(Fig.3)”; all tables as ’’Table 3” or ’’(Table 3)”.

Galley Proofs & Reprints

The corresponding author will receive proofs in PDF format prior to the
final publishing of the manuscript. Corrections to the proofs must be
returned to the Managing Editor within five days; failure to return these
corrections promptly will result in delay of publication. The Academy will
provide 100 reprints without charge for each feature article or note published
in the Journal. These will be mailed to the corresponding author or the
designated contact person following the publishing of each issue of the
Journal, The distribution of reprints among coauthors is the responsibility of
the corresponding author. Authors will have the opportunity to purchase
additional reprints (in lots of 100) at the time that the corrected galley proofs
are returned to the Managing Editor.

Voucher Specimens

When appropriate, such as new records, noteworthy range extensions, or
faunal or floral listings for an area, the author(s) should provide proper
information (to include accession numbers) relative to the deposition of
voucher specimens. Specimens should be placed with the holdings of a

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 1, 2007

recognized regional or national museum or herbarium. The name(s) and
designated initials used by the museum should be given as part of the
introduction or methods section. Do not site the deposition of voucher
specimens in personal collections.

The Editorial Staff is very aware that many members of the Academy
work with organisms that are protected by state or federal regulations. As
such, it may not be possible to collect nor deposit these specimens as
vouchers. In the interest of maintaining credibility, authors are expected to
provide some alternate means of verification such as black and white
photographs, list of weights or measurements, etc. The Editorial Staff
retains the option to determine the validity of a record or report in the
absence of documentation with a voucher specimen.

Page Charges

Authors are required to pay $50 per printed page. While members of the
Academy are allowed four published pages per year free of charge on
publications, all authors with means or institutional support are requested to
pay full page charges. Full payment is required for all pages in excess of
four. Nonmembers of the Academy are required to pay full page charges for
all pages. The Academy, upon written request, will subsidize a limited
number of pages per volume. These exceptions are, however, generally
limited to students without financial support. Should a problem arise
relative to page charges, please contact:

Dr. Ned E. Strenth
TJS Managing Editor
Department of Biology
Angelo State University
San Angelo, Texas 76909
E-mail: ned.strenth@angelo.edu

Additional Guidelines

An expanded version of the above author guidelines which includes
instructions on style, title and abstract preparation, deposition of voucher
specimens, and a listing of standardized abbreviations is available on the
Academy's homepage at:

www.texasacademyofscience.org

THE TEXAS ACADEMY OF SCIENCE, 2006-2007

OFFICERS

President
President Elect
Vice-President:

Immediate Past President
Executive Secretary:
Corresponding Secretary:
Managing Editor:

Manuscript Editor:

Treasurer:

AAAS Council Representative:

David S. Marsh, Angelo State University

Hudson R. DeYoe, University of Texas-Pan American

Raymond C. Mathews, Jr., Texas Water Dev. Board

Damon E. Waitt, Lady Bird Johnson Wildflower Center

Fred Stevens, Schreiner University

Cindy Contreras, Texas Parks and Wildlife Department

Ned E. Strenth, Angelo State University

Frederick B. Stangl, Jr., Midwestern State University

James W. Westgate, Lamar University

Sandra S. West, Southwest Texas State University

DIRECTORS

2004 Benjamin A. Pierce, Southwestern University

Donald L. Koehler, Balcones Canyonlands Preserve Program

2005 Jerry L. Cook, Sam Houston State University
Flo Oxley, Lady Bird Johnson Wildflower Center

2006 Herbert D. Grover, Hardin-Simmons University
Gary P. Garrett, Texas Parks and Wildlife Department

SECTIONAL CHAIRPERSONS

Anthropology: Roy B. Brown, Instituto Nacional de Antropologia y Historia

Botany: Randall G. Terry, Lamar University

Cell and Molecular Biology: Ana Christensen, Lamar University

Chemistry and Biochemistry: Robert Holloway, Schreiner University

Computer Science: John A. Ward, Brook Army Medical Center

Conservation Ecology: Diana Comuzzie, Schreiner University

Environmental Science: Forrest M. Mims III, Seguin, Texas

Freshwater and Marine Sciences: Brian W. Brooks, Baylor University

Geosciences: Christian O. George, University of Texas at Austin

Mathematics: William D. Clark, Stephen F. Austin State University

Physics: David Bixler, Angelo State University

Science Education: Kaycie Sullivan, Texas Tech Campus at Junction

Systematics and Evolutionary Biology: Allan W. Hook, St. Edward’s University

Terrestrial Ecology and Management: Caren McLemore, Weatherford, Texas

COUNSELORS

Collegiate Academy: William J. Quinn, St. Edward's University
Junior Academy -.NmcQ Schielack, Texas A&M University

PERIODICALS

THE TEXAS JOURNAL OF SCIENCE
Texas Academy of Science
CMB 6252
Schreiner University
Kerrville, Texas 78028-5697

iiMi ,

TEXAS JOURNAL

SCIENCE

PUBLISHED QUARTERLY BY
THE TEXAS ACADEMY OF SCIENCE

Volume 59
Number 2
May 2007

GENERAL INFORMATION

www.texasacademyofscience.org

Dr. Fred Stevens, Exeeutive Seeretary
The Texas Academy of Science
CMB 6252
Schreiner University
Kerrville, Texas 78028-5697
E-mail: FStevens@schreiner.edu

THE TEXAS JOURNAL OF SCIENCE

Volume 59, No. 2

May, 2007

CONTENTS

Concerning Triangle Preserving Functions.

By John W. Spellmann and Ricardo M. Torrejon . . . . . 83

An Analytical Investigation to Improve the Buckling Capacity of
Hollow Square Pipe Columns.

By M. A. Faruqi and P. K. Shah . . . . . . . . . . 97

Characterization of a High Energy Macroalgal Community in Quintana Roo,
Mexico Using Digital Image Analysis.

By Ryan L. Pikes, Leslie C. Smith and Roy L. Lehman . . . 103

Seasonal Variation in Dune Vegetation at South Padre Island, Texas.

By Frank W. Judd, Robert L Lonard, K. Rod Summy

and Ruben A. Mazariegos . 1 1 3

Effects of Temperature and Illumination on Background Matching in
Mediterranean Geckos (Hemidactylus turcicus).

By Frederic Zaidan III and Pamela L. Wiebusch . 127

A Comparison of Mammalian Diversity Across Vegetational Associations of
the Aramberri Region of Nuevo Leon in Northeastern Mexico.

By Jessica Valero-Padilla, Armando J. Contreras-Balderas,

Jose Ma. Torres- Ayala and Salvador Contreras-Arquieta . 137

General Notes

Aggressive Head-Up Displays in Great-Tailed Grackles
{Quiscalus mexicanus).

By Elissa S. Wampler and David E. Gammon . 151

Second Report of the Southern Painted Turtle, Chrysemys dorsalis
(Testudines: Emydidae), from Texas, with Comments on its Genetic
Relationship to Other Populations.

By Chris T. McAllister, Michael R. J. Forstner
and Jonathan P. Fuller

155

THE TEXAS JOURNAE OF SCIENCE
EDITORIAL STAFF

Managing Editor:

Ned E. Strenth, Angelo State University
Manuscript Editor:

Frederick B. Stangl, Jr., Midwestern State University
Associate Editor for Botany:

Janis K. Bush, The University of Texas at San Antonio
Associate Editor for Chemistry:

John R. Villarreal, The University of Texas-Pan American
Associate Editor for Computer Science:

Nelson Passos, Midwestern State University
Associate Editor for Environmental Science:

Thomas LaPoint, University of North Texas
Associate Editor for Geology:

Ernest L. Lundelius, University of Texas at Austin
Associate Editor for Mathematics and Statistics:

E. Donice McCune, Stephen F. Austin State University
Associate Editor for Physics:

Charles W. Myles, Texas Tech University

Manuscripts intended for publication in the Journal should be submitted in
TRIPLICATE to:

Dr. Frederick B. Stangl, Jr.

TJS Manuscript Editor
Department of Biology
Midwestern State University
Wichita Falls, Texas 76308
frederick.stangl@mwsu.edu

www.texasacademyofscience.org

AFFILIATED ORGANIZATIONS
American Association for the Advancement of Science,

TEXAS J. OF SCI. 59(2):83-96

MAY, 2007

CONCERNING TRIANGLE PRESERVING FUNCTIONS

John W. Spellmann and Ricardo M. Torrejon

Department of Mathematics
Texas State University-San Marcos
San Marcos, Texas 78666

Abstract -The concept of a triangle preserving function is introduced and it is
proved that functions of the form f{x) = belong to this class for integers

k>2 and real numbers 0 < < A: .

Conditions under which three arbitrary positive real numbers
a,b, and c may serve as the length of the sides a triangle has been
the subject of inquiry for some time. Arguably, the oldest of the
necessary and sufficient conditions is the cyclic inequality

a + b-c>0,b + c-a>0,c + a-b>0, (1)

from which further necessary and sufficient conditions have been
derived. An argument that used the following reformulation of (1)

+b^ +c'^)< a^b^ +b^c^ +c^a^ (2)

lead V. E. Hoggat, Jr. (1959) to proposed the following problem:

Show that if a,b,c form a triangle, then fa,fb,4c form a
triangle.

Together with Hoggafs own proof, two other solutions
extending his result were also published: R.T. Hood's (1960) proof
which replaces the square roots with roots of an arbitrary order n ,
and J. L. Brown's proof (1960) where these roots are replaced with
a nonnegative nondecreasing subadditive function / defined on
(0,0)).

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

The search for sufficient conditions implying the feasibility of
triangles that are based on the measurement of the angles of a given
triangle naturally leads to triangle preserving functions.

For this paper a triangle preserving function (TPF) is a
continuous function / \{0,7r)-^R^ which satisfies the property that
for any triangle T with interior angles a,p and y , there exists a
triangle r* whose sides have lengths f{a),f{P) and f{y). Nothing
is implied about triangles T* and T being the same or different.

T. P. Cerpanova in (1963; 1966) studied the functions
/(x) == cos(x/2) and /(x) = cos^(x/2). She showed that each function
was a TPF. In this paper the authors expand this work by showing
that functions of the form f{x) = cos {xlk) are TPFs for integers k >2
and real numbers 0 < 5 < ^ .

Background

In this section a number of theorems will be presented which
will serve as background for the main results of this paper. First,
from elementary geometry one has that a necessary and sufficient
condition for three line segments to be the sides of a triangle is that
the sum of the lengths of any two of the segments must be greater
than the length of the third line segment. This is the first theorem:

Theorem 1. Suppose that / is a continuous function.

Then f will be a TPF if and only if each of the following
inequalities hold for any triple a,p and y of positive numbers
whose sum is n :

0 f{cc)<f{P)vf{y)\

ii) /(/?)</(6^) + /(r);and

iii) f{Y)<f{cx) + f{P).

SPELLMANN & TORREJON

Note 1. Whenever convenient and without loss of generality
one assumes that if is a continuous function and if

a, p and Y are the interior angles of a triangle, then the angles have
been labeled so that the inequality f{a)> f{P)> f(r) holds. In this
case one needs only to show that f(P) + fir) > ficc) in order to prove
that / is a TFP.

An immediate consequence of Theorem 1 is that positive
constant functions are TPFs.

Theorem 2. Suppose k>0 and fix) = k for 0<x<;r. Then f is
a TPF.

Proof For any triangle with interior angles a, p and one has
that fia)^ fiP) = fiy) = k which are the lengths of the sides of an
equilateral triangle with sides of length k . □

Before continuing the authors would like to note that there are
continuous functions / : (0,;r) ^ which are not TPFs.

Example 1. Consider the function defined by /(x) = jc^ and a
30-60-90 degree triangle. Letting a = nil, p = tzP and y = nl6, one
sees that it is not true that /(«)< f(P) + fir) since ;rV4> ;rV9 + ;rV36.
Thus, the function fix) = is not a TPF.

Also, as TPFs must be positive valued, one notes that

Example 2. The function /(x) = cos(x/^) defined on (0,;z-) is not a
TPF for any real number 0<k<2.

Theorem 1 and the linear nature of summation give that TPFs
easily can be combined to produce other TPFs. This is given in the
next theorem.

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

Theorem 3. Suppose that each of is a TPF, k>0

and each of is a positive constant. Then the function

F defined on (0,7r) by the rule

F(x) = k + Yc<,f.{x)

(=1

is a TPF.

At this point it is recalled that a function is sub¬

additive if f(x + y)< f{x) + f{y) for all in the domain of /. The
function will be super-additive if f{x + y)> f{x) + f{y) for all x,y in
the domain of / . Results related to this paper for sub-additive and
super-additive functions may be found in Hoggat (1959; 1960) and
Brown (1960). In particular, the next two theorems are noted.

Theorem 4. Suppose that f\R^^R^ is increasing and sub¬
additive. If a, b and c are lengths of sides of a triangle, then
f{a),f{b) and /(c) will be lengths of sides of a triangle.

Theorem 5. Suppose that f:R*^R^ is increasing and super¬
additive. If a, b and c are positive numbers which can't be the
lengths of the sides of any triangle, then no triangle exists whose
sides have lengths f {a), f {b) and f (c) .

The next theorem notes that power functions ( /(x) = x" for 5 > 0 )
provide good examples of sub-additive and super-additive
functions.

Theorem 6. Suppose that .s>0 and f is the power function
defined by f (x) = x^ for x > 0 . Then

i) / is sub-additive if 0 < .s < 1 ; and

SPELLMANN & TORREJON

ii) / is super-additive if s>\.

Proof It is noted that the proof is straight forward for 5 = 0 or 1
(when f{x) = \ or /(x) = x). Thus one considers that either 0 < 5 < 1
or s>\. Let H be the function defined by H{t) = {\ + ty -f -\ for
/>0.

It is noted that + Thus, H’{t)>0 for ^>1 and

H'{t)<t) for 0<5<1. Since H(0) = 0, it follows that H is positive
valued for ^ > l and negative valued for 0 < 5 < 1 .

Now, for positive x and y one has that

f{x + y)-f{x)-f(y) = (x + yY-x^~/

From this one sees that / is sub-additive when 0<5<l and

super-additive when 5 > 1 .

□

The next theorem uses Theorem 6 to show that certain powers of
TPFs are themselves TPFs while certain powers of non-TPF
functions will not be TPFs.

Theorem 7. i) Suppose f:{0,7i)^R^ is a TPF and 0<5<1.
Then f is a TPF.

ii) Suppose f :(0,7r)^R^ is not a TPF and s>\. Then f is
not a TPF.

Proof i). For ^ = 0, the function f" will be the constant function
at one which is a TPF by Theorem 2. Now consider the case where

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

0<s<l. Suppose that a,fi and / are the interior angles of a
triangle. Since / is a TPF, then f{a),f{P) and /(/) will be the
lengths of the sides of some triangle. Since, for 0<.s<l, the
function H(x) = x' is sub-additive. Theorem 6 then gives that
H(f{a)) = riaXH(f(P))-r(P) and H(f(r)) = r(r) will be the sides
of a triangle which means that /" is a TPF.

ii). Since / is not a TPF there exists a triangle with interior
angles a,p and y such that f{a),f{P) and /(/) will not be the
lengths of the sides of any triangle. Now, for .^>1, the function
H(x) = x' is super-additive. It then follows from Theorem 6 that
H{fia)) = ria),H(f(P)) = r(P) and H{f{y)) = r{y) will not be the
sides of any triangle. Thus, /' will not be a TPF. □

With this background the authors now shift to the results of this
paper which are given in the theorems that follow.

Main Results

As previously noted, Cerepanova (1963; 1966) proved that each
of the functions /(x) = cos(jc/2) and cos^(x/2) is a TPF. In this paper
the authors extend these results to show that functions of the form
cos' {xlk) are TPFs for integers k>2 and real numbers
0<s<k.

One starts by showing that the function f (x) ^ cos^^ {x/2k) will be
a TPF whenever the function /(x) = cosH^/^) is a TPF.

Theorem 8. Suppose k>2 is a positive integer and the function
f{x)^cosfxlk) defined on (O./r) is a TPF. Then the function
f (x) = cos^^ {xl2k) for 0 < X < ;r is a TPF. Moreover, functions f of
the form f (x) - cos' {xl2k) for t)<s<2k will be TPF.5.

Proof Since the function /(x) = cosH^/^) is a TPT, it follows
form Theorem 7 that f{P)^ cos\xlk) is a TPF for / = l,2,...,^. Using

SPELLMANN & TORREJON

the cosine half-angle formula gives that

m = cos^v

^x/k^

COS

— + — cos
_2 2

^kj

— V —
2* "^^2*

COS I —

/(Xf

The fact that the function f {x) = cos^\xl2k) is a TPF then follows
from Theorem 3. Using the fact that the function /(x) = cos^Hx/2^) is
a TPF and the fact that 0 < sl2k < 1 , Theorem 7 may then be used to
show that functions / of the form /(x) = cosTx/2A:) are TPFs for

^<s<2k. □

Theorem 9* Suppose that k>2 is an integer. Then the function
f defined on (0,;r) by f{x) = coi{xlk) is a TPF.

Proof The proof is now broken into four cases: ^ = 2,3,4, and

k>5.

Case 1: k = 2. This, as previously noted, was proved by
C erapanova in (C erapanova 1963; C erapanova 1966).

Case 2: A: = 3 . For this part one notes the triple angle formula

cos(3x) = 4cos^(x)-3cos(x). (3)

Replacing x with x/3 and cos(x) with 2cos^(x/2)-l allows one to
rearrange (3) to give

/x^

3 f

1 2

"■"I

= — cos

+ -COS

4 1

90 THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

Now suppose that a, and / are the interior angles of a triangle
labeled so that f{a)> f{P)> /{y). To eomplete this case, it suffices
to show that f{P) + f{r)>f{a) or cos^(y^/3) + cosTy3)-cos^(6?/3) > 0 .
Now

cos

, !>(

1 Gc)

+ cos —

“ cos -

V 3 7

> U

(p) (

f O')

COS

— - cos

+ —

UJ 1

5JJ

(4)

cos

^ I , 21/1 2

Using Case 1 ( /(x) = cos^ (x/2) is a TPF) and the fact that
0 < cos(«/3) < 1 , statement (4) may be altered to give

fp'

[

fy)

+ COS

“COS TT r T

cos —

+ cos -

k3j

1 [3 J 4

V 3 y

[3)]

1. (5)

As / is a strictly decreasing function one notes that
0<a< p<Y<7r . Also, that a + p + 71 . Using these two facts gives
that 0 < y^/3 < 7d6 and 0 < ^3 < ;z/3 . Thus,

cos — > —

UJ 2

and cos

n>i

(6)

Using the inequalities (6) in (5) gives that

cos

Trees U -cos I

> —

S 1

- 1 -

2 2

1>0.

(7)

From (7) it follows that /(x) = cos^(x/3) is a TPF and the proof of
this case is complete.

Case 3: /r = 4 . This follows from Case 1 and Theorem 8.

SPELLMANN & TORREJON

Case 4: k>5. With k>5, consider the function f{x) = coi{xlk)
for 0<x<;r. Suppose that a,p and y are the interior angles of a
triangle labeled so that f(a)> f{p)>f{Y). To complete this case
one needs only to show that /(A) + /(r) > /(«) or that
co^\pik) + cos\ylk)> co^''{alk). From the definition of / one then
sees that

a < P <Y < K.

one concludes that

a 4- /? + 7 = ^,

7 < TT and p < nil.

(8)

An elementary calculus argument shows that each of {cos^ (.mX-,
and (cos* (#)}r.5 is a strictly increasing positive sequence in k which
converges to 1 . This fact, together with (8) yields

Thus,

This completes Case 4 and the proof of the theorem. □

Having proved that fimctions of the form f{x) = coi{xlk) are
TPFs for integers k>2, the authors will now expand this result to
include all functions of the form f{x) = cos {xlk) for any integer k>2
and any real number ^ , with 0 < ^ < /c .

Theorem 10* Suppose that k>2 is an integer and that s is a

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

real number satisfying 0<s<k. Then the function f{x) = cos{xlk)
defined on {g.n) is a TPF.

Proof For 5 = 0, / is the constant function at one which means
that / is a TPF using Theorem 2. For 0<s<k one notes that
0<^/A:<l. By Theorem 7 the function fy'\x)=^cos{xlk) is a TPF
which completes the proof of this result. □

It is noted for integers k>2 and numbers s>k that functions of
the form f{x)^cos{x/k) sometimes are TPFs but not always. This is
illustrated by Theorems 12 and 13. Theorem 11 is a lemma to
Theorem 12.

Theorem 11.

(a) For 0 < X < ;r , the sequence

an increasing sequence; and

(b) ^ ^ integer n>\.

Proof (a) Let 0<x<;r and ^>1. The cosine double angle
formula and repeated use of the algebraic difference of squares
formula gives that

> 0.

SPELLMANN & TORREJON

This shows that the sequence is increasing.

(b) This part follows using the fact that each term of an
increasing sequence is greater than or equal to the first term of the
sequence.

Theorem 12. Suppose n is a positive integer and 0<5<2^”“'.
Then the function f defined on (0,;r) by the rule /(x) = cosT^/2") is

a TPF.

Proof The authors will first prove that the function g defined
on (0,;r) by the rule g(x) = cos^^'' *(x/2”) is a TPF. Let af and / be
the interior angles of a triangle labeled so that g{a)> g{P)> g^y) . To
show that g ia a TPF one needs only to show that g(/?) + g{y) > g(a) .
Now, using Theorem 1 1, it may be seen that

/ \

r /

^ \

1 1

4U-1

I I

1 ^ 1

2 (

— -

cos

cos -

- cos

U”J

[r)\

L UJJ

gix) = cos^

for 0 < X < ;z" . Using statement (9) and the trig identity

cos

(/^] ifr]

= 2 + 2sin

■ (fi) ■

(r]

+ COS

[2 J+cos [-J

sm — sm

[2)

one then notes that

giP) + gir) = + g(P) + gir) - 2g(or) + g(a)

cos

= 2 + 2sin
> g(a).

A ^1

+ COS

— + cos

^ 1

^2 J

•

{p '\ ■

(r]

[2)

{2)

I2J

22n-\

- ^ COS

COS

- l + gia)

(9)

Thus, the function g is a TPF. The fact that the function / is a TPF

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

will then follow from Theorem 7 and the fact that 0 < < 1 , since

_ 5/2"

Theorem 13. Suppose that s>2 and /(x) = cosT^/2) for
0<x<7r . Then f is not a TPF.

Proof. For 5 > 2 , it is noted that

liml + (l + sin(p)r^^ =1 + 1 = 2

and that

lim(l + cos(/.)U = 2*'^>2.

Thus, let 0 < /? < tt/A be a number such that

l + (l + sin(p)y'' <(l + cos(/?)F'^ (10)

Using the fact that sm{p) = cos(;t/2-p), statement (10) gives

Thus,

1 + cos(;z/2 - /?)

+ cos(/?)

COS^

+ cos^

7112- p^

This means that if r is a triangle with interior angles ;z/2, 7il2-p and
p, then there will not exist a triangle with sides of length
f(7r/2),(^2-p) and /(/?). Thus / is not a TPF. □

Theorem 14. Suppose k>2 is an integer and p is a real
number such that the function f{x) = cos^(x/^) defined on (0,7r) is not
a TPF. Then for s> p, the function g{x) = cos" {xlk) is not a TPF.

SPELLMANN & TORREJON

Proof. This theorem follows from Theorem 7 as slp>\ and

gW = W-

□

It has been asked (see Aassila (2005)) if functions of the form
/(x) = sinT^/2) would be TPFs for ^>0. The next theorem answers
this question in the negative.

Theorem 15. Suppose ^>0 and the function f is defined on
(0,;r) by the rule /(x) = sinTx/2) . Then f is not a TPF.

Proof For fixed ^ > 0 , one notes that

Now let 0 < p < ;r/4 be such that

0<sin^ ^ ~<cosT/^)<F

V 2 y 4 2

(11)

Let T be the triangle with interior angles

K -Ip and P = y = p.

(12)

Then, using (11) and (12), have that

cosT/^)

(13)

From statement (13) see that for 5>0, functions of the form

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

/ (x) = sin'(x/2) cannot be TPFs.

Conclusion

In this paper the authors have extended the results of
Cerepanova (1963; 1966) and have shown that functions of the
form f(x) = cos{xlk) are TPFs for integers k>2 and real numbers
0<s<k. What is not done and what is conjectured by the authors
(based on theorems contained in this paper as well as other
numerical evidence) is that for each integer k>2, there exits a
number s{k) such that the function / defined on (0,;r) by the rule
/ (x) = cos {xlk) is a TPF if and only if 0 < .9 < s{k) .

Literature Cited

Aassila, A. 2005. Problem 1717, Mathematics Magazine, 78(2):158.

Brown, J. L., Jr. 1960. Solution IV to problem E 1366, American Mathematical
Monthly, 66(1):83.

Cerepanova, T. P. 1963. Matematika v skole, 3:89.

Cerepanova, T. P. 1966. Matematika v skole, 6:66.

Hoggat, V. E., Jr. 1959. Problem E 1366, American Mathematical Monthly,
66(5):423.

Hoggat, V. E., Jr. 1960. Solution II to problem E 1366, American Mathematical
Monthly, 66(1 ):82.

Hood, R. T. 1960. Solution to problem E 1366, American Mathematical Monthly,
66(1): 83.

JWS at: js20(gtxstate.edu

TEXAS J. SCI. 59(2):97-102

MAY, 2007

AN ANALYTICAL INVESTIGATION TO IMPROVE THE BUCKLING
CAPACITY OF HOLLOW SQUARE PIPE COLUMNS

M. A. Faruqi and P. K. Shah

MSC 194

Department of Civil Engineering
Texas A&M University - Kingsville
Kingsville, Texas 78363

Abstract.-This study examined the effects of pressurization on buckling capacity
in hollow square pipes. Analytical relationships between different parameters
involved in this process are presented. Results reveal that pressurization not only
increases the buckling capacity of hollow square pipes but can also reduce the
amount of required material.

Pressurizing hollow square pipes has never been examined as a
means of increasing the buckling capacity. The main goal of this
paper is to analytically show the effectiveness of pressurizing
hollow square pipe columns as a means of improving their buckling
capacity and saving material. Analytical relationships between dif¬
ferent parameters in the buckling analysis are presented. It was
found that pressurizing can increase the buckling capacity as well
as help in saving material.

A number of investigations (Jahsman 1964; Wilhoit & Merwin
1967; Reddy & Calladine 1978 ) have been conducted to improve
the buckling capacity of pipe columns using internal pressure.
However, these investigations were limited to thin circular pipes.
Further, the researchers were mainly interested in predicting the
wrinkling effects of such pipes. In this work, an elastic buckling
analysis of pressurized hollow square pipe columns is presented.
Buckling is of prime importance in many structures where pipes are
used as columns. Pipe pressurizing involves introducing air
through a hole in the pipe which has both ends sealed. This method
can be continued until the hoop (circular) stress reaches an
allowable tensile stress.

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

Figure 1. Cross-section of a hollow pressurized square column.

Analytical Analysis

The generalized column under investigation is tubular in shape,
has a length (L, ft), has a constant moment of inertia (I, in"^), is
made of linearly elastic material, and has a constant elastic modulus
(E, ksi). The column is subjected to an axially applied compressive
load. Please note that in this analytical investigation English units
are being used. However, international systems of units (SI) can
equally be used. Figure 1 shows the cross-section of a hollow
pressurized square column.

The load carrying capacity of a regular circular column is given
by Euler’s equation:

P,, = Tt^Er^A/L^ (1)

Per = Maximum load carrying capacity of a column, kips
A = Cross-sectional area of the column, in^

r = Radius of gyration, in

FARUQI & SHAH

The maximum shear flow capacity (Higdon et. al 1985) through a
hollow circular section is as follows:

shear flow * pressurized area = (2)

For a hollow pressurized pipe (Figure 1), equation 1 can be written
as:

Per = load carried by the material + load carried by the internal
pressure, therefore

(3)

p = Internal pressure in the pipe, ksi
r = Radius of gyration of the shaded area =>

Im = Moment of inertia of the shaded area, in"^

Am = Area of the shaded material, in^

rp = Radius of gyration of the pressurized section

Ip = Moment of inertia of the pressurized section, in"^
Ap = Area of the pressurized section, in

t = Thickness of the column, in

Substituting the area values corresponding to shaded and
pressurized part of a hollow square column with width (b, in), we
have:

-{b-2tY\ + ^[b-2t)

(4)

100

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

Figure 2. Critical Load vs. Width for a Typical Thickness, t = 0.375 in.

Figure 3. Critical Load vs. L/t for a Typical Width, b = 9 in.

Calculations and Comparison of Results

The results presented in this paper are derived from the
analytical equation above. The maximum load carrying capacity of
pressurized and non-pressurized pipes of various heights are
evaluated. Typical heights (12, 16, and 24 ft), thickness (0.375 in),
width (9 in) and corresponding pressures of 0 ksi and 2 ksi with
each typical height are substituted into equation 3. This evaluation
is shown in Figures 2 and 3. These figures show comparative

FARUQI & SHAH

101

Table 1. Comparison of critical loads for randomly selected parameters.

Width (Inches)

(P crw/Pcro)

L= 12ft

L= 16ft

L = 24ft

4.5

1.08

1.15

1.33

6.5

1.09

1.16

1.37

L/t

(P crw/P cro)

200

1.02

1.01

1.02

600

1.25

1.13

1.18

*Pcrw/ Pcro = cHtical loads with internal pressure/critical loads without internal pressure

Table 2. Comparison

of Critical Loads With Amw/Amo.

Per (kips)

(Amw /Amo)

L= 12 ft

L= 16 ft

L = 24ft

0.97

0.95

0.9

750

0.97

0.95

0.89

1000

0.97

0.95

0.89

*Aniw/Amo = area of pipe material with internal pressure/area of pipe without internal pressure

critical pressure plots with various design parameters. It is
noticeable from Figures 2 and 3 that the critical load carrying
capacity of pressurized columns increases with increase in width
and thickness. Also, for a particular critical design pressure a
smaller area is required for a pressurized pipe. A typical
comparison summary is shown in Tables 1 and 2. These results can
respectively translate into increasing the buckling (load carrying)
capacity and saving material (thinner pipes).

Conclusions and Recommendation

The main goal of this paper is to analytically show the
effectiveness of pressurizing hollow square pipe columns as a
means of improving their buckling capacity and saving material.

The results from this limited analytical study show that,
pressuring a pipe before using it as a column is very effective in
developing tensile stress and thus increasing its buckling capacity.

102

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

Also, a non-pressurized pipe supporting a compressive axial load
can be replaced by a thinner pressurized pipe. This can translate
into substantial savings of material and money. However,
additional work is required to develop design standards and
guidelines to ensure its safe and economical use.

Literature Cited

Jahsman, W. E. 1964. The behavior of a pressurized circular cylindrical membrane
column under combined bending and compression. Lockhead Technical Report:
6-114.

Wilhoit, J. C. & Merwin, J. E. 1967. Pipe stresses induced in laying offshore
pipeline. J. Industrial Engineering, 89:37-46.

Reddy, B. D. & Calladine, C. R. 1978. Classical buckling of a thin walled tube
subjected to bending moment and internal pressure. International Journal of
Mechanical Science,20(9): 641-650.

Higdon, A., E. Ohlsen, W. Stiles, J. Weese & W. Riley. 1985. Mechanics of
Materials, Wiley, New York, 744 pp.

MAE at: M-Faruqi@tamuk.edu

TEXAS J. SCL 59(2): 103- 112

MAY, 2007

CHARACTERIZATION OF A HIGH ENERGY MACRO ALGAL
COMMUNITY IN QUINTANA ROO, MEXICO USING DIGITAL
IMAGE ANALYSIS

Ryan L. Fikes, Leslie C. Smith
and Roy L. Lehman

Center for Coastal Studies, Texas A&M University-Corpus Christi
6300 Ocean Drive, Corpus Christi, Texas 78412

Abstract.-Macroalgal community structure was characterized for a high energy
portion of the lower midlittoral, upper infralittoral zone of a rocky point on the
eastern coast of Mexico’s Yucatan Peninsula. This intertidal zone of Punta Yu Yum
has historically been inaccessible due to high energy wave action. However,
meteorological conditions during a 2005 summer research expedition made it
possible to conduct a one-day intensive sampling event. A species checklist was
developed for the macroalgal community, and three sets of voucher specimens were
collected of each species encountered. A transect consisting of 154 quadrats (20 cm
by 30 cm), imaged with a Sony 4.1 megapixel Cyber-Shot® digital camera, was used
to collect data along the shoreline. Images were evaluated with Coral Point Count
(CPCe), which assigned 25 random points per image. At each point the alga was
identified to lowest possible taxon. Data from all quadrats was analyzed for species
diversity, richness, evenness, and relative cover. In total, 10 orders of marine
macroalgae were represented. Chondrophycus papillosa (Laurencia papillosa) had
the greatest relative cover at 37.08%, followed by Turbinaria tricostata with 17.64%
cover. Species Richness was 31, with classes Florideophyceae constituting 10
species, Phaeophyceae 9 species, and Chlorophyceae 10 speeies. The maximum
Shannon Diversity Index was found to be 1.51 (mean of 0.97), following normal
trends of low diversity in high energy habitats. This study supports sampling by
means of digital imagery as an effective tool in data collection.

Numerous studies on zonation have been conducted on rocky
shores around the world (Raffaelli & Hawkins 1996). The eastern
shoreline of the Yucatan is largely composed of these shores in the
form of limestone outcroppings. These environments are generally
high in energy, resulting in low biological diversity towards the
infralittoral ridge (Raffaelli & Hawkins 1996). The soft ironshore
limestone of this area contains little or no iron and takes its name
from the characteristic resonance similar to that of an anvil (Britton
& Morton 1989). There have been relatively few studies conducted
on the benthic marine algae associated with these Caribbean rocky
ironshores.

104

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

Intertidal communities are strongly influenced by both
biological and physical processes (Sutherland & Ortega 1986). A
universal zone system (Stephenson & Stephenson 1972) was
proposed for rocky shores which divide it into three major zones:
supralittoral fringe (high shore), midlittoral zone (midshore), and
infralittoral fringe (low shore). The term universal is meant to be
taken in the sense that there are features, or trends, that are
recurrent and widespread. Tides, wave action, climate, topography,
and substrate are all factors that contribute to this intertidal zonation
(Taylor 1978). The goal of this project was to characterize the high
energy, rarely exposed, portion of the lower midlittoral, upper
infralittoral zone on a rocky point on the eastern coast of Mexico's
Yucatan peninsula.

Punta Yu Yum is located south of the city of Tulum, Mexico,
within the Sian Ka'an Biosphere Reserve (Figure 1). It is one of
many rocky outcroppings occuring along a sand bar peninsula,
which extends south from the mainland at the northern boundary of
the reserve. The point consists of heavily weathered limestone with
a gentle slope. The high shore and midshore zones are easily
accessible, but the intertidal, low shore zone lies within an area of
extreme energy. According to Milliman (1973), this region is
tropical as mean rainfall exceeds 1 00 cm"^*^ and mean annual air and
seawater temperatures range from 26°-28°C. In addition to
dominant northeasterly trade winds causing easterly currents, this
area is also characterized by semi-diurnal tides rarely exceeding
amplitudes of 0.5- 1.0 m (Milliman 1973).

Materials and Methods

The lower portion of the low shore zone of Punta Yu Yum was
surveyed on the morning of 24 May 2005. The area experienced
exceptionally low tides during late May due to the unusual path of
hurricane Adrian passing over El Salvador into the Caribbean Sea.
The algal community was surveyed by capturing 154 images with a
Sony 4.1 megapixel Cyber-Shot® digital camera. Images were
taken at random intervals along a belt-like transect, running parallel

FIKES, SMITH & LEHMAN

105

.Ujiifefd Statesi'

^Bahama

Jslands

Is la Mujeres

Gulf of Mexico

«/Sinto Pupno
Dctniiicio Rico

South America

Punta Yu Yum

Sian Ka.'ai
Biosphen
Reserve.

C^gintana

Lake -
Bacalai^!^ /

Belize r

Punta

Cancun

r^Punta

'•'nIzuc

Puerto

Morelos

Ascencion

Bay

Punta Piedra

Espirto Santo
Bay

Punta Herrero

Chinchorro

Bank

Caribbean

Sea

Figure 1. Map showing study site (Punta Yu Yum) in relation to Sian Ka’an Biosphere
Reserve and the Yucatan Penninsula. Modified from Tunnell et al. (1993).

to the shoreline. A copper- tubing quadrat (20 cm by 30 cm) was
used to maintain consistency in sample (image) size, and each
quadrat was placed along the exposed high-energy area of the rocky
ironshore. The digital image was taken so that the image border
matched that of the quadrat frame. Representative algal specimens
were collected from sampling sites, pressed for herbarium purposes.

106

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

and archived at Texas A&M University-Corpus Christi. All
material was identified with the aid of Littler & Littler (2000).

Images were analyzed with Coral Point Count software with
Excel extensions (Kohler & Gill 2006). CPCe is a Windows-based
program that provides a tool for the determination of eoral eover
and diversity using transect photographs and the random point
count method. For this study 25 random points per image were
seleeted and the alga at that point identified to the lowest possible
taxon. Coral code files were modified in the CPCe program by
replacing coral species with algal species. This transformation
allows for the analysis of benthic algal communities. Data from all
quadrats was analyzed for speeies diversity, richness, evenness, and
relative cover using CPCe and Quantitative Analysis in Ecology
software programs.

Results

Of the 154 images taken, 132 (n) were usable for habitat
characterization, giving a total of 3,300 points for all quadrats. The
remaining photographs were not focused correctly, making it
impossible to verify species eontent. Species richness was 3 1 taxa,
with the classes Florideophyceae constituting 10 species,
Phaeophyceae 9 speeies, and Chlorophyceae 10 species (Table 1).
Chondrophycus papillosa [previously Laurencia papillosa] had the
greatest relative cover with 37.08%, followed by Turbinaria
tricostata with 17.64% and Cladophora spp. with 12.80% cover
(Figure 2). Red macroalgae made up the largest percent
composition of the study with 46.95% (Table 2). In total, 10 orders
of marine macroalgae were represented. Maximum Shannon
Diversity was found to be 1.51, with an Evenness of 0.59. Mean
Shannon Diversity for all samples was 0.97 (± 0.02 SE).

Discussion

Results of this study follow normal trends of low diversity in
high energy habitats. In 1999, the macroalgae at three roeky
outeroppings in the Sian Ka'an Biosphere Reserve, ineluding Punta

FIKES, SMITH & LEHMAN

107

Table 1. Taxonomic list showing species encountered in quadrat samples at Punta
YuYum, Quintana Roo, Mexico, May 2005. Nomenclature follows Wynne (2005).

Rhodophyta
Eurhodophytina
Class: Florideophyceae

ORDER: BONNEMAISONIALES
Family: Bonnemaisoniaceae

Asparogopsis taxiformis (Delile) Trevis.

ORDER: CERAMIALES

Family: Rhodomelaceae

Digenea simplex (Wulfen) C. Agardh

Chondrophycus papillosus (C. Agardh) Garbary & J.T. Harper

ORDER: CORALLINALES
Family: Corallinaceae

Subfamily: Corallinoideae

Jania spp. J.V. Lamour
Subfamily: Lithophylloideae

Amphiroa spp. J.V. Lamour

ORDER: GELIDIALES

Family: Gelidiellaceae

Gelidiella acerosa (Forssk.) Feldmann & Hamel

ORDER: GRACILARIALES
Family: Gracilariaceae

Gracilaria spp. Grev.

Hydropuntia crassissima (P. Crouan & H. Crouan) M.J. Wynne

Ochrophyta
Class: Phaeophyceae

ORDER: DICTYOTALES

Family: Dictyotaceae

Dictyota spp. J.V. Lamour

Lobophora variegata (J.V. Lamour) Womersley

Padina sanctae-crucis Borgesen

ORDER: FUCALE

Family: Sargassaceae

Sargassum fluitans (Borgesen) Borgesen
Sargassum hystrix J. Agardh
Sargassum natans (L.) Gallon

Sargassum polyceratium (var. ovatum) (Collins) W.R. Taylor
Turbinaria tricostata E.S. Barton
Turbinaria turbinata (L.) Kuntze

108

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

Chlorophyta
Class: Chlorophyceae

ORDER: BRYOPSIDALES
Eamily: Caulerpaceae

Caulerpa racemosa (Forsskal) J. Agardh
Caulerpa spp. J.V. Lamour

ORDER: CLADOPHORALES
Family: Cladophoraceae

Cladophora spp. Kiitz
Family: Halimedaceae

Halimeda spp. J.V. Lamour
Family: Siphonocladaceae

Cladophoropsis macromeres W.R. Taylor
Dictyosphaeria cavernosa (Forssk.) Borgesen
Ventricaria ventricosa (J. Agardh) J.L. Olsen & J.A. West
Family: Valoniaceae

Valonia macrophysa Kiitz

ORDER: DASYCLADALES
Family: Dasycladaceae

Dasycladus vermicularis (Scop.) Krasser

Yu Yum, were compared (Albert & Lehman 1999). This study
surveyed the low shore and midshore zones of Punta Yu Yum and
described the richness to be 15 species. Species diversity was
highest for Chlorophyta (8 species), followed by Rhodophyta (4
species) and Phaeophyta (3 species). Kolterman (2000) surveyed
four rocky points within the Sian Ka’an Reserve, including Punta
Yu Yum, with vertical transects. Of the four locations, Punta Yu
Yum had the greatest species richness (20), with Digenia simplex,
Dictyota spp., and Laurencia spp. dominant. Tunnell et al. (1993)
described the rocky ironshore at Punta Chahuay, Quintana Roo,
Mexico as having nine species in the “brown zone” (high energy
zone). Similar to the current study, dominant species reported
included Sargassum polycarpum, Turbinaria tricostata, and
Cladophoropsis macromeres.

Results of this study showed an increase in species richness to
3 1 when compared to these previous studies, with the Rhodophyta
exhibiting the greatest number of species. Algal species richness.

FIKES, SMITH & LEHMAN

109

Blue-green Algae
Cladophoropsis macromeres
Cladophora spp.
Turbinaria tricostata
Sargassum polyceratium
Condrophycus pa pi 1 1 os a
Hydropuntia crassissima
Gracillaria spp.
Digenea simplex

10 20 30 40

Relative Cover (%)

Eigure 2. Relative cover of dominant algal species identified for Punta Yu Yum in
southern Quintana Roo, Mexico, May 2005. Percent coverage is for all images (132)
in transect. Nomenclature follows Wynne (2005).

however, for the Sian Ka’an reef system in its entirety has been
reported as 212, with Oehrophyta dominant (Navaro & Robinson
1990; Keeney 1999). Though speeies riehness for this site has
inereased with this study, it is important to note that it is
comparable to other reef zones. Keeney (1999) described the
species richness for reef zones as follows: lagoon/seagrass beds
(30), patch reefs (62), backreef (43), shallow reef (36), and mid to
deep reef (36). Zones with a higher richness have lowered
disturbance in the form of wave energy.

Disturbance has dramatic effects on macroalgal communities.
Herbivory is the major biotic disturbance for reef systems, and the
removal of herbivores often results in rapid overgrowth of
macroalgae (Liddell & Ohlhorst 1986; Sammarco 1982). In high
energy environments, herbivory is reduced, allowing for rapid
growth of algae. Herbivory defense mechanisms, such as
halogenated metabolites in Chondrophycus (Fenical 1975), CaCOs
in Turbinaria, and tannins in Sargassum, further allow macroalgae
to thrive in this high energy habitat.

110

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

Table 2: Percent coverage for transect and Shannon Diversity for each major category
(group) observed at Punta Yu Yum, Quintana Roo, Mexico, May 2005. Percent
coverage is for all images (132) in transect.

Major Category

Percent of

Transect

Shannon

Diversity

Red Macroalgae - Rhodophyta

46.95

0.77

Brown Macroalgae - Ochrophyta

26.18

0.99

Green Macroalgae - Chlorophyta

18.35

0.96

Calcareous Algae

1.36

Other Live - Cyanobacteria

2.11

0.14

Sand, Pavement, Rubble

4.12

Unknowns

0.92

Tape, Wand, Shadow

11.00

In rocky shore communities, moving water can rip macroalgae
off of the substrate, causing major physical disturbance (Koehl &
Wainwright 1985). From a functional form perspective, dominant
species for this study encompassed the coarsely branched-group
and thick leathery-group, those most resistant to grazers (Littler et
al. 1983). This functional form model also measures “resiliency” of
species, primarily in response to filament penetration. Species of
the thick lethery-group, such as Sargassum polyceratium, withstood
up to 909 g-cm^ surface pressure, far higher than those of the
filamentous group (<200 g-cm^ surface pressure) (Littler et al.
1983). This “resiliency” may translate into ability to withstand
pounding waves, such as in the lower midlittoral, upper infralittoral
zone of Punta Yu Yum.

This study provides additional data on this algal community with
its inclusion of the typically inaccessible lower intertidal
community. It is important to characterize these habitats in their
entirety, allowing ecologists to better understand habitat form and
function in the ecosystem. Global biodiversity is dependent on
worldwide censuses, allowing for comparative analysis (Liddell &
Ohlhorst 1991). This study also demonstrates that sampling by
means of digital imagery may prove successful in habitat
characterization, producing comparable results to destructive

FIKES, SMITH & LEHMAN

111

sampling. Programs such as CPCe may be especially helpful in
studies which are time-limited or require collecting permits.

Acknowledgments

Thanks to Dr. John “Wes” Tunnell, Jr. of the Harte Research
Institute for Gulf of Mexico Studies for making it possible to
conduct a field expedition to the Caribbean. Thank you to everyone
with the Sian Ka’an Biosphere Reserve for continuing to welcome
our university to this truly unique region of Quintana Roo. Thanks
to Dr. Henry Hildebrandt for aiding in the funding to make this trip
possible and to Kevin Kohler for assistance with the CPCe
program.

Literature Cited

Albert, E. M. & R. L. Lehman. 2000. Marine algae associated with Caribbean rocky
shores, Quintana Roo, Mexico. J Phycol., 37:4-4.

Britton, J. C. & B. Morton. 1989. Shore ecology of the Gulf of Mexico. University
of Texas Press, Austin, Texas. 387 pp.

Fenical, W. 1975. Halogenation in the Rhodophyta - a review. J. Phycol., 11: 245-
259.

Keeney, T. S. 1999. Coral reef macroalgae in northern Sian Ka’an Biosphere
Reserve, Quintana Roo, Mexico. M. S. Thesis, Texas A&M University-Corpus
Christi, 58 pp.

Koehl, M. A. R. and S. A. Wainwright. 1985. Biomechanics. Pp. 291-313, in
Handbook of phycological methods. Ecological Field Methods: Macroalgae
(M.M. and D.S. Littler, eds.), Cambridge University Press, 617 pp.

Kohler, K. E. & S. M. Gill. 2006. Coral Point Count with Excel extensions (CPCe):
A Visual Basic program for the determination of coral and substrate coverage
using random point count methodology. Comput & Geosci 32(9): 1259-1269.

Kolterman, A. 2000. Ecological characterization of northwestern Caribbean
ironshores, Quintana Roo, Mexico. M. S. Thesis, Texas A&M University-
Corpus Christi, Corpus Christi, Texas, 82 pp.

Liddell, W. D. & S. L. Ohlhorst. 1986. Changes in benthic community composition
following the mass mortality of Diadema at Jamaica. J. Exp. Mar. Biol. Ecol.,
95:271-278.

Littler, D. S. & M. M. Littler. 2000. Caribbean reef plants: An identification guide
to the reef plants of the Caribbean , Bahamas, Florida, and Gulf of Mexico.
Offshore Graphics, Inc., Washington, DC, 542 pp.

Littler, M. M., D. S. Littler & P. R. Taylor. 1983. Evolutionary strategies in a
tropical barrier reef system: functional-form groups of marine macroalgae. J.
Phycol., 19: 229-237.

112

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

Milliman, J. D. 1973. Caribbean coral reefs. Pp. 1-50 in O. A. Jones and R. Endean,
Eds. Biology and geology of coral reefs, Vol. 1: geology. Academic Press Inc.,
New York, 456 pp.

Navarro, L. D. & J. G. Robinson (eds.). 1990. Diversidad Biologica en la Reserva
Biosfera de Sian Ka’an, Quintana Roo, Mexico. CIQRO, 471 pp.

Raffaelli, D. & S. Hawkins. 1996. Intertidal Ecology. Chapman and Hall, London,
England. 356 pp.

Sammarco, P. W. 1982. Effects of grazing by Diadema antillarum Philippi
(Echinodermata: Echinoidea) on algal diversity and community structure. J. Exp.
Mar. Biol. EcoL, 63:83-105.

Stephenson, T. A. & A. Stephenson. 1972. Life between tidemarks on rocky shores.
W. H. Freeman and Company, San Francisco, California, 435 pp.

Sutherland, J. P. & S. Ortega. 1986. Competition conditional on recruitment and
temporary escape from predators on a tropical rocky seashore. J. Exp. Mar. Biol.
EcoL, 95:155-156.

Tunnell, J. W., A. A. Rodriguez, R. L. Lehman & C. R. Beaver. 1993. An ecological
characterization of the southern Quintana Roo coral reef system. Center for
Coastal Studies, TAMU-CC - 9307 - CCS, Texas A&M University-Corpus
Christi. Corpus Christi, Texas, 161 pp.

Wynne, M. J. 2005. A checklist of benthic marine algae of the tropical and
subtropical western Atlantic: second revision. Gebruder Bomtraeger, Berlin,
Germany, 152 pp.

RLF at: ryan.fikes@tamucc.edu

TEXAS J. OF SCI. 59(2): 1 13-126

MAY, 2007

SEASONAL VARIATION IN DUNE VEGETATION
AT SOUTH PADRE ISLAND, TEXAS

Frank W. Judd, Robert I. Lonard, K. Rod Summy
and Ruben A. Mazariegos*

Department of Biology and "^Department of Physics and Geology
University ofTexas-Pan American
Edinburg, Texas 78541

Abstract.-Seasonal variation in vegetative cover, species richness, species
composition and species importance were compared in the backshore and primary
dunes topographic zones at a dune protection area and an unprotected site on South
Padre Island, Texas, from May 2004 to April 2005. Cover was greatest in fall, but
there was no significant variation among seasons in percent cover in the backshore or
primary dunes at either site. Species richness was significantly greater at the dune
protection site, but there was no significant seasonal variation in species richness in
either topographic zone at either site. Species composition differed between
topographic zones and sites, but was similar among seasons within a topographic
zone at a given site. Heterotheca subaxillaris was either the dominant species or a
co-dominant species in both the backshore and primary dunes at the dune protection
site. There was significant variation in dominance in the backshore and primary
dunes at the unprotected site. Ipomoea pes-caprae was the dominant species in warm
seasons and Croton punctatus was dominant in winter.

Padre Island is the longest of the barrier islands of the Texas
eoast. It extends 182 km southward from Corpus Christi to Brazos-
Santiago Pass at Port Isabel. The southern third of the island is
separated from the northern two-thirds by the Mansfield Channel.
This southern segment of the island is known as South Padre Island,
and the name also is applied to an ineorporated town that oeeupies
the southern 8 km of the island. South Padre Island is 55 km long
and has an area of about 16,200 ha (Lonard et al. 1999).

The flora and vegetation of South Padre Island are relatively
well known eompared to other islands of the Texas eoast (Dahl et
al. 1974; Judd et al. 1977; Lonard et al. 1978; Lonard & Judd 1980;
Lonard & Judd 1981; Judd & Sides 1983; Judd & Lonard 1985;
Judd & Lonard 1987; Judd et al. 1989; Lonard & Judd 1989; Judd
et al. 1991; Lonard et al. 1991; Everitt et al. 1991; Everitt et al.

114

THE TEXAS JOURNAL OF SCIECE-VOL. 59, NO. 2, 2007

1992; Lonard & Judd 1993; Judd et al. 1994; Lonard & Judd 1997;
Judd et al. 1998; Lonard & Judd 1999; Everitt et al. 1999; Lonard et
al. 1999; Summy et al. 2006), but except for the information
provided by Lonard & Judd (1989) on flowering phenology, there
are no data on seasonal variation in the vegetation. Sampling
protocols in a study of the relationship between vegetative cover
and elevation change provided an opportunity to assess seasonal
variation in percent cover, species richness, species composition
and species importance at two sites on South Padre Island which are
presented herein.

Materials and Methods

Based on ground observations and an aerial survey, two study
sites were selected that appeared to differ markedly in vegetation
abundance. Site 1 is in a "dune protection zone" where driving and
pedestrian traffic in the dunes is prohibited. Vegetation at this site
appeared to be abundant and diverse. The site is located at 26° 14'
34" N and 97° 17' 20" W. It included a broad area of secondary
dunes and vegetated flats. Site 2 is about 7.4 km north of Site 1 at
26° 18' 37" N and 97° 17' 39" W in an unprotected area. It appeared
to have markedly less vegetation than Site 1 and included only a
narrow strip of secondary dunes and vegetated flats.

At each site, three parallel transects spaced 25 m apart were
established extending 90° E from the margin of Highway 100 to the
high tide line at the Gulf of Mexico shore. To facilitate relocation
of transects, the starting points at Highway 100 were recorded using
a survey grade GPS unit and marked with orange paint. The
transects passed through three topographic zones and part of a
fourth zone (Judd et al. 1977), i.e. foreshore, backshore, primary
dunes, and part of the secondary dunes and vegetated flats. The
foreshore is bare. It is the portion of the Gulf of Mexico shore over
which the tide migrates daily. This study focused on seasonal
changes in vegetation of the backshore and primary dune zones.

JUDD ET AL.

115

Vegetation was sampled at four periods: 1) early summer - 30
May to 5 June 2004, 2) mid fall - 5 to 12 October 2004, 3) mid
winter - 27-30 January 2005 and 4) mid spring - 6 to 9 April 2005.
The line-intercept method (Canfield 1941) was used to quantify
vegetation abundance and distribution. Each transect was divided
into 10 m intervals and data were recorded along the total length of
each interval. First, the total cover of live and standing dead
vegetation intercepted by the line was determined. Then, each live
species intercepted by the line was rated individually and recorded
without separation into strata. Frequency and foliage cover were
recorded for each species and from these data relative frequency,
relative cover and an importance value that was the sum of relative
frequency and relative cover were calculated. Importance values
were used to determine dominance. The topographic zone of each
interval was recorded.

Results

Comparison of mean annual percent cover between the study
sites showed that the visual inspection was correct. Cover was
significantly greater in both the backshore and primary dunes at
Site 1 (Table 1). There was no significant seasonal variation in
percent cover in the backshore or primary dune zones at either site
(Table 1) (backshore Site 1, F - 1.701; # = 3, 8; F - 0.244;
primary dunes Site 1, F = 2.083; df= 3, S; P 0.180; backshore
Site 2, F = 3.052; df= 3, 8; F == 0.092; primary dunes Site 2, F =
1.845; df= 3, 8; F = 0.217). Cover was highest in fall in each of
the topographic zones and at each site. Cover was lowest in winter
in both the backshore and primary dunes at Site 2, but lowest in
spring in both zones at Site 1 .

Species richness was greater at Site 1 than at Site 2 (Table 2),
especially in the backshore zone. There was no significant seasonal
variation in species richness in either of the topographic zones at
either of the sites (Table 2).

116

THE TEXAS JOURNAL OF SCIECE-VOL. 59, NO. 2, 2007

Table 1. Comparison of mean annual percent cover and seasonal variation in percent
cover of live and standing dead vegetation at two sites and in two topographic zones
at each site on South Padre Island, Texas. Site 1 is in a dune protection zone, and
Site 2 is in an unprotected area. N = sample size, SD = standard deviation of the
mean.

Season

Site 1 Backshore % Cover

N Mean SD

Site 2 Backshore % Cover

N Mean SD

Summer 04

58.48

5.51

4.11

0.95

Fall 04

62.36

4.37

6.99

4.10

Winter 05

56.36

5.14

1.17

1.66

Spring 05

54.40

2.44

3.06

1.68

Annual Mean

57.90

4.86

3.84

3.01

t (annual mean) = 32.747, 22 df,P < 0.001

Season

Site 1 Primary Dunes % Cover

Site 2 Primary Dunes % Cover

Mean

Summer 04

45.73

5.24

23.76

12.55

Fall 04

57.86

7.72

32.26

13.12

Winter 05

50.05

12.21

14.29

8.56

Spring 05

42.47

4.37

16.38

5.36

Annual Mean

49.03

9.09

21.67

11.52

t (annual mean) = 9.237, 22 df, P < 0.001

Table 2. Comparison of seasonal variation in species richness in the backshore and
primary dunes topographic zones at two sites on South Padre Island, Texas. Sum =
summer, Wn = winter, Spr = spring.

Site

Backshore

X" Test

Primary Dunes

X“ Test

Sum

Fall

Spr

Sum

Fall

Spr

1.203,3 #

0.388, 3 df

P>0.5

P>0.9

0.273, 3 df

1.163,3#

P>0.9

P>^.5

There was no seasonal variation in dominance among seasons in
the backshore zone at Site 1. Heterotheca subaxillaris (camphor
weed) was the dominant species in each season (Table 3). Species

JUDD ET AL.

117

Table 3. Comparison of species importance in the backshore zone at Site L Cover is of
live plants. Freq. == frequency, Rel. Freq. = relative frequency, Rel. Cover = relative
cover, IV = importance value.

Season

Species

Freq.

Rel.

Freq.

Cover

Rel.

Cover

Summer 04

Heterotheca subaxillaris

46.7

33.4

19.14

41.2

74.6

Uniola paniculata

18.0

12.7

6.04

12.9

25.6

Oenothera drummondii

16.7

12.0

5.00

10.5

22.5

Chamaecrista fasciculata

11.4

8.6

4.41

10.1

18.7

Indigophera miniata

13.8

10.0

3.62

7.9

17.9

16 additional species, sum of their IVs =

40.7

Fall 04

Heterotheca subaxillaris

41.4

29.6

17.36

35.0

64.6

Chamaecrista fasciculata

22.8

16.3

10.92

22.0

38.3

Oenothera drummondii

15.7

11.0

5.53

11.1

22.1

Uniola paniculata

10.5

7.5

3.91

7.9

15.4

Indigophera miniata

6.4

4.9

2.18

4.0

8.9

12 additional species, sum of their IVs =

50.7

Winter 05

Heterotheca subaxillaris

40.7

43.8

12.97

58.2

102.0

Oenothera drummondii

13.3

14.1

3.23

14.4

28.5

Uniola paniculata

11.3

12.2

3.05

13.6

25.8

Ipomoea pes-caprae

7.9

8.5

1.02

4.6

13.1

Chamaecrista fasciculata

5.6

5.9

0.93

3.8

9.7

10 additional species, sum of their IVs =

20.9

Spring 05

Heterotheca subaxillaris

39.4

38.6

12.76

49.7

88.3

Uniola paniculata

15.3

15.5

5.85

21.8

37.3

Chamaecrista fasciculata

13.3

12.9

1.33

5.4

18.3

Oenothera drummondii

5.9

5.8

1.95

7.2

13.0

Ipomoea pes-caprae

6.1

6.0

1.42

5.8

11.8

1 1 additional species, sum of their IVs =

31.3

ranking 2 through 5 in importance varied among seasons, but
Uniola paniculata (sea oats), Oenothera drummondii (beach
evening primrose), and Chamaecrista fasciculata (partridge pea)
were among the top five speeies in importance in each season.
Consequently, four out of five of the top five speeies were eommon

118

THE TEXAS JOURNAL OF SCIECE-VOL. 59, NO. 2, 2007

Table 4. Comparison of species importance among seasons in the backshore zone at Site
2. Cover is of live plants. Abbreviations as in Table 3.

Season

Species

Freq.

Rel.

Freq.

Cover

Rel.

Cover

Summer 04

Ipomoea pes-caprae

24.4

81.3

5.58

83.8

165.1

Ipomoea imperati

4.4

14.7

0.98

14.7

29.4

Croton punctatus

1.2

2.6

0.10

1.5

5.5

Fall 04

Ipomoea pes-caprae

21.6

74.5

4.36

65.6

140.1

Croton punctatus

7.0

24.1

2.28

34.3

58.4

Sesuvium portulacastrum

0.4

1.4

0.01

0.3

1.5

Winter 05

Croton punctatus

4.0

47.6

0.98

83.1

130.7

Ipomoea pes-caprae

3.8

45.2

0.19

16.1

61.3

Ipomoea imperati

0.6

7.1

0.01

0.8

7.9

Spring 05

Ipomoea pes-caprae

7.0

51.5

1.29

48.7

100.2

Croton punctatus

6.6

48.5

1.36

51.3

99.8

to each season so that species composition was similar among
seasons.

There was never more than three species present in the
backshore zone at Site 2 and in Spring 2005 only two species were
present (Table 4). Ipomoea pes-caprae (railroad vine) was the
dominant species in summer and fall and a co-dominant in spring.
Croton punctatus (beach croton) was the dominant species in
winter. Thus, there was seasonal variation in dominance. Camphor
weed was not present in the backshore at Site 2.

Camphor weed was a clear dominant in the primary dune zone at
Site 1 in summer, winter, and spring, and it was a co-dominant with
partridge pea in fall (Table 5). Thus, dominance was relatively
stable among seasons. Three species (camphor weed, partridge pea,
and sea oats) were among the top five species in importance in each
season. These three species were also present in the top five
species in each season in the backshore zone at Site 1.

JUDD ET AL.

119

Table 5, Comparison of species importance among seasons in the primary dune zone at
Site 1. Cover is of live plants. Abbreviations as in Table 3.

Season

Species

Freq.

Rel.

Freq.

Cover

Rel.

Cover

Summer 04

Heterotheca subaxillaris

39.1

30.6

15.05

48.9

69.5

Chamaecrista fasciculata

32.2

24.8

9.52

24.0

48.8

Uniola paniculata

24.5

19.1

6.55

17.4

36.5

Croton punctatus

5.2

4.4

1.19

3.4

7.8

Indigophera miniata

2.6

2.0

1.75

4.7

6.7

1 additional species, sum (

3f their IVs =

30.7

Fall 04

Chamaecrista fasciculata

49.3

30.8

23.79

38.8

69.6

Heterotheca subaxillaris

43.2

26.8

21.93

35.9

62.7

Uniola paniculata

27.7

17.2

6.79

11.5

28.7

Ipomoea pes-caprae

7.3

4.4

2.14

3.4

7.8

Indigophera miniata

3.3

2.1

1.04

1.5

3.6

9 additional species, sum of their IVs =

27.6

Winter 05

Heterotheca subaxillaris

48.4

37.5

14.88

54.8

92.3

Uniola paniculata

26.5

20.1

4.82

17.8

37.9

Chamaecrista fasciculata

20.0

16.3

1.81

6.7

23.0

Oenothera drummondii

5.2

3.8

1.99

7.3

11. 1

Schizachyrium littorale

7.7

6.0

1.17

4.3

10.3

8 additional species, sum of their IVs =

25.4

Spring 05

Heterotheca subaxillaris

42.6

35.2

10.27

47.0

82.2

Chamaecrista fasciculata

28.8

23.2

4.12

19.3

42.5

Uniola paniculata

22.0

18.1

3.90

18.1

36.2

Schizachyrium littorale

9.0

7.6

1.31

7.1

14.7

Croton punctatus

1.7

1.6

0.42

2.0

3.6

1 additional species, sum of their IVs ==

20.8

Consequently, speeies eomposition of the backshore and primary
dune zones was similar at Site 1 .

There was seasonal variation in dominance in the primary dune
zone at Site 2. Railroad vine was the dominant species in summer
and fall and beach croton was the dominant in winter and spring

120

THE TEXAS JOURNAL OF SCIECE-VOL. 59, NO. 2, 2007

Table 6. Comparison of species importance among seasons in the primary dune zone at
Site 2. Cover is of live plants. Abbreviations as in Table 3.

Season

Species

Freq.

Rel.

Freq.

Cover

Rel.

Cover

Summer 04

Ipomoea pes-caprae

34.9

33.2

9.17

34.2

67.4

Croton punctatus

24.8

23.6

8.01

29.9

53.5

Ipomoea imperati

21.9

20.8

2.91

10.9

31.8

Panicum amarum

9.8

8.3

0.89

3.3

11.6

Heterotheca subaxillaris

4.7

3.4

1.77

6.6

10.0

5 additional species, sum

of their IVs =

25.7

Fall 04

Ipomoea pes-caprae

39.3

37.0

11.10

37.5

74.5

Croton punctatus

22.9

18.9

10.05

24.7

43.6

Ipomoea imperati

23.9

19.0

4.60

12.2

31.2

Heterotheca subaxillaris

7.8

5.7

6.10

14.9

20.6

Panicum amarum

16.4

12.9

1.99

5.2

18.1

4 additional species, sum of their IVs =

12.0

Winter 05

Croton punctatus

33.1

44.5

8.08

63.0

107.5

Ipomoea pes-caprae

12.1

15.6

1.42

11.1

26.7

Ipomoea imperati

12.7

16.4

0.98

7.6

24.0

Panicum amarum

11.4

13.8

0.88

6.0

19.8

Heterotheca subaxillaris

3.9

5.2

2.24

16.5

19.7

3 additional species, sum of their IVs =

2.3

Spring 05

Croton punctatus

18.9

20.4

4.91

32.8

53.2

Ipomoea pes-caprae

18.9

20.4

2.94

19.6

40.0

Panicum amarum

15.3

17.3

1.41

9.4

26.7

Heterotheca subaxillaris

6.7

6.9

2.35

15.7

22.6

Ipomoea imperati

11.0

12.3

1.03

6.9

19.2

4 additional species, sum

of their IVs =

38.3

(Table 6). Conversely, speeies eomposition was similar among
seasons. The same five speeies: railroad vine, beach croton,
Ipomoea imperati (beach morning glory), Panicum amarum (bitter
panicum), and camphor weed comprised the top five species in
importance in each season.

JUDD ET AL.

121

Discussion

It was expected that peak cover would occur in fall. Greatest
precipitation typically occurs in September (Tunnell 2002). About
20.5% (14.0 cm) of the annual rainfall (68.2 cm) occurs in this
month and 32.6% of the annual rainfall occurs in September and
October combined. Thus, vegetative cover was expected to be
greatest after peak rainfall and while temperature remains high, i.e.,
during fall.

Because most species are perennials, it was not surprising that
species richness did not vary significantly among seasons. Site 1
had greater species richness and cover than has been reported
previously. Judd et al. (1977) reported species richness, percent
cover and species importance in the backshore and primary dune
zones of South Padre Island based on 18 transects taken at 3.2 km
intervals along the length of the island. They found that species
richness was eight species in the backshore zone and that total
cover of live vegetation was 4.7%. The annual mean total cover for
Site 2 was only slightly lower than their value, but species richness
at Site 2 was markedly lower (three) than the eight species they
found. Species richness and percent cover values in the backshore
in the dune protection zone at Site 1 were 2.2 and 12.3 times
greater, respectively, than the values reported by Judd et al. (1977)
for the backshore. Clearly, the curtailment of vehicular traffic in
the backshore zone increased species richness and vegetation
abundance.

The same conclusion holds true for the primary dunes. Judd et
al. (1977) reported 11 species in the primary dunes and cover of
7.1% for the windward slope and 21.3% for the leeward slope of
the primary dunes. Species richness in the primary dunes at Site 1
varied among seasons from 11 to 14 and cover was 49.03%, which
is 3.5 times greater than that reported by Judd et al. (1977).

122

THE TEXAS JOURNAL OF SCIECE-VOL. 59, NO. 2, 2007

Similarity of species composition among seasons at a given site
and topographic zone is likely a consequence of most species being
perennials. Seasonal variation in dominance in the backshore and
primary dune zones at Site 2 appears to reflect differences in
response to cold by railroad vine and beach croton. On December
25, 2004, subfreezing temperature (-2.0°C) occurred at Brownsville,
Texas (20 km southwest of the study area) and a rare snowfall of
about 3.8 cm blanketed the area. Apparently, beach croton is more
tolerant to cold and retains more of its leaves, whereas railroad vine
loses many of its leaves in winter. There are no data on the
physiological responses to cold for either species, but the
geographic distribution of the species supports the contention that
beach croton is more tolerant of cold. Beach croton reaches North
Carolina in its northward distribution (Correll & Johnston 1970)
while railroad vine does not extend north of southernmost Georgia
(Devall 1992). Furthermore, Devall (1992) reports that railroad
vine is limited to tropical and subtropical zones between 30°N and
30°S latitudes and that, in some areas, the above-ground portions of
the plant die off in winter leaving underground stems to sprout
when conditions again become favorable.

There have been changes in species composition and importance
since Judd et al. (1977) first reported on the vegetation of South
Padre Island. Judd et al. (1977) found that sea oats and Sesuvium
portulacastrum (sea purslane) were of almost equal importance in
the backshore, but these two species occurred in two distinct belts.
Sea purslane occurred in a belt closest to the high tide line and sea
oats in a belt on foredunes in advance of the primary dunes. The
top five species in the backshore and their importance values were:
sea oats (71.9), sea purslane (70.0), beach morning glory (22.6),
bitter panicum (1 1.8) and railroad vine (10.7). Species composition
and importance at Site 2 was closer to the composition and
importance reported by Judd et al. (1977) than was Site 1. A belt of
sea purslane was not present at either Site 1 or 2. Judd and Sides
(1983) reported the belt of sea purslane was obliterated by

JUDD ET AL.

123

Hurricane Allen in 1980. Lonard et al. (1999) showed that sea
purslane had not recovered by 1997 and that species richness in the
backshore was the same (three species) 17 years later as it was
immediately after Hurricane Allen. Lonard et al. (1999) suggested
that the decrease in species richness and relative importance of sea
oats was caused by vehicular damage to vegetation and dunes in the
backshore zone.

Judd & Sides (1983) reported that vegetation of the primary
dunes was not greatly affected by Hurricane Allen. Species
richness and percent cover were similar in pre- and post-hurricane
transects. However, they found a shift in dominant species after the
hurricane. Prior to Hurricane Allen the dominant species on both
the windward and leeward slopes of the primary dunes was beach
morning glory and sea oats was second in importance on both
slopes. After the hurricane, sea oats was dominant on the windward
slopes of the primary dunes and beach morning glory was second in
importance. On the leeward slope of primary dunes, Schizachyrium
littorale (seacoast bluestem) was dominant in post-hurricane
transects and sea oats was second in importance.

Species composition and importance in the primary dunes at Site
2 were more similar to species composition and importance
reported in the primary dunes by Judd et al. (1977), Judd & Sides
(1983) and Lonard et al. (1999) than was species composition and
importance at Site 1. A major difference at Site 2 was the relative
importance of railroad vine and beach croton. Railroad vine ranked
fifth in importance in pre-hurricane transects on the windward
slopes of the primary dunes and fourth in importance on the
leeward slopes (Judd et al. 1977; Judd & Sides 1983). Croton
punctatus ranked fourth on windward slopes and fifth on leeward
slopes. In the present study, the relative importance of these two
species had increased to first and second (order varied among
seasons).

124

THE TEXAS JOURNAL OF SCIECE-VOL. 59, NO. 2, 2007

In summary, there was no significant seasonal variation in
species richness or percent cover in either of the topographic zones
at either of the sites. Species composition was similar among
seasons in both topographic zones at both sites, but there was
seasonal variation in which species was dominant in the primary
dunes at Site 1 and in both topographic zones at Site 2. Seasonal
variation in dominance appears to reflect differences in response to
cold by railroad vine and beach croton with railroad vine being
more sensitive to cold and losing many of its leaves in winter.

Acknowledgments

We thank the National Oceanic and Atmospheric Administra¬
tion, the Coastal Coordination Council of the Texas General Land
Office, and the University of Texas-Pan American for financial
support.

Literature Cited

Canfield, R. H. 1941. Application of the line interception method in sampling range
vegetation. Journal of Forestry 39 (4):388-394.

Correll, D. S. & M. C. Johnston. 1970. Manual of the vascular plants of Texas.
Texas Research Foundation, Renner, Texas, xv + 1-1 88 1pp.

Dahl, B. E., B. A. Fall, A. Lohse & S. G. Appan. 1974. Stabilization and
reconstruction of Texas foredunes with vegetation. Gulf Universities Research
Consortium 139:1-325.

Devall, M. S. 1992. The biological flora of coastal dunes and wetlands. 2. Ipomoea
pes-caprae (L.) Roth. Journal of Coastal Research, 8 (2):442-456.

Everitt, J. H., D. E. Escobar & F. W. Judd. 1991. Evaluation of airborne video
imagery for distinguishing black mangrove {Avicennia germinans) on the lower
Texas gulf coast. Journal of Coastal Research 7 (4): 1 169-1 173.

Everitt, J, H., D. E. Escobar, F. W. Judd & M. R. Davis. 1992. Evaluation of spot
satellite and airborne video imagery for distinguishing black mangrove
{Avicennia germinans). Pp. 169-175, in Proceedings 13th Biennial Workshop on
Color Aerial Photography and Videography in the Plant Sciences, Orlando,
Florida. May 6-9, 1991, (Published in 1992).

Everitt, J. H., M. A. Alaniz, D. E. Escobar, R. 1. Lonard, F. W. Judd & M. R. Davis
1999. Reflectance characteristics and film image relations among important plant
species on South Padre Island, Texas. Journal of Coastal Research, 15(3):789-
795.

JUDD ET AL.

125

Judd, F. W., R. L Lonard & S. L. Sides. 1977. The vegetation of South Padre Island,
Texas in relation to topography. Southwestern Naturalist, 22(1):3 1-48.

Judd, F. W. & S. L. Sides. 1983. Effects of Hurricane Allen on the nearshore
vegetation of South Padre Island. Southwestern Naturalist, 28(3):365-369,

Judd, F. W. & R. 1. Lonard. 1985. Effects of perturbations on South Padre Island.
Pp. 1855-1869, in Proceedings Fourth Symposium on Coastal and Ocean
Management, "Coastal Zone *85". American Soc. of Civil Engineers. Baltimore,
Maryland, 2672 pp.

Judd, F. W. & R. 1. Lonard. 1987. Disturbance and community development. Pp.
1731-1745, in Proceedings Fifth Symposium on Coastal and Ocean Management,
“Coastal Zone ’87". American Soc. of Civil Engineers. Seattle, Washington,
5870 pp.

Judd, F. W., R. I. Lonard, J. H. Everitt & R. Villarreal. 1989. Effects of vehicular
traffic in secondary dunes and vegetated flats of South Padre Island, Texas. Pp.
4634-4645, in Proceedings Sixth Symposium on Coastal and Ocean Management,
"Coastal Zone '89", American Soc, of Civil Engineers. Charleston, South
Carolina, 4978 pp.

Judd, F. W., R. I. Lonard, J. H. Everitt, & D. E. Escobar. 1991. Resilience of
seacoast bluestem barrier island communities. Pp. 3513-3524, in Proceedings
Seventh Symposium on Coastal and Ocean Management, "Coastal Zone ’91".
American Soc. of Civil Engineers. Long Beach, California, 3800 pp.

Judd, F. W., R. 1. Lonard, D. L. Hockaday, D. E. Escobar, J. H. Everitt & R. Davis.
1994. Remote sensing of nearshore vegetation. South Padre Island, Texas. Pp.
581-589, in Proceedings of the Second Thematic Conference on Remote Sensing
for Marine and Coastal Environments. Vol. I. Needs, Solutions and Applications.
Environmental Research Institute of Michigan (ERIM), P. O. Box 134001, Ann
Arbor, MI 48113-4001, U.S.A., 704 pp.

Judd, F. W., R. I. Lonard, J. H. Everitt, D. E. Escobar & M. R. Davis. 1998. Wind-
tidal flats and dune vegetation of South Padre Island, Texas. Pp. 177-184, in Vol.
II, Proceedings Fifth International Conference on Remote Sensing for Marine and
Coastal Environments. San Diego, California, 5-7 October 1998, 584 pp,

Lonard, R. I., F. W. Judd & S. L. Sides. 1978. Annotated checklist of the flowering
plants of South Padre Island, Texas. Southwestern Naturalist, 23(3):497-5 10.
Lonard, R. I. & F. W, Judd. 1980. Phytogeography of South Padre Island, Texas.
Southwestern Naturalist, 25(3):3 13-322.

Lonard, R. 1. & F. W. Judd. 1981. The terrestrial flora of South Padre Island, Texas.

Texas Memorial Museum Miscellaneous Papers No. 6, 74 pp.

Lonard, R. I. & F. W. Judd. 1989. Phenology of native angiosperms of South Padre
Island, Texas. Pp. 217-222, in Proceedings Eleventh North American Prairie
Conference. T. Bragg and J. Stubbendieck (Eds.). Univ. Nebraska Printing,
Lincoln, Nebraska, 293 pp.

Lonard, R. L, F. W. Judd, J. H. Everitt, & D. E. Escobar. 1991. Roadside associated
disturbance on coastal dunes. Pp. 2823-2836, in Proceedings Seventh Symposium
on Coastal and Ocean Management, "Coastal Zone '91". American Soc. of Civil
Engineers. Long Beach, California, 3800 pp.

126

THE TEXAS JOURNAL OF SCIECE-VOL. 59, NO. 2, 2007

Lonard, R. I. & F. W. Judd. 1993. Recovery of vegetation of barrier island washover
zones. Pp. 2324-2331, in Proceedings of the Eighth Symposium on Coastal and
Ocean Management, "Coastal Zone '93". American Soc. Civil Engineers. New
Orleans, Louisiana, 3512 pp.

Lonard, R. I. & F. W. Judd. 1997. The biological flora of coastal dunes and
wetlands. Sesuvium portulacastrum (L.) L. Journal of Coastal Research, 13
(1):96-104.

Lonard, R. I. & F. W. Judd. 1999. The biological flora of coastal dunes and
wetlands. Ipomoea imperati (Vahl) Griseb. Journal of Coastal Research,
15(3):645-652.

Lonard, R. L, F. W. Judd, J. H. Everitt, D. E. Escobar, M. A. Alaniz, I. Cavazos III &
M. R. Davis. 1999. Vegetative change on South Padre Island, Texas, over
twenty years and evaluation of multispectral videography in determining
vegetative cover and species identity. Southwestern Naturalist, 44(3):261 -27 1 .

Summy, K. R., F. W. Judd, R. I. Lonard, R. A. Mazariegos, J. H. Everitt & M. R.
Davis. 2006. Effectiveness of aerial color-infrared photography for mapping
vegetative cover within dune areas of South Padre Island, Texas. Proceedings of
the 20‘^ Biennial Workshop on Aerial Photography, Videography, and High
Resolution Digital Imagery for Resource Assessment, Weslaco, Texas. October
4-6, 2005. American Society of Photogrammetry and Remote Sensing, 12 pp,
CD ROM (unpaged).

Tunnell, Jr., J. W. 2002. The Environment. Pp. 73-84, in Tunnell, Jr., J. W. and F.
W. Judd (eds.). The Laguna Madre of Texas and Tamaulipas. Number Two:
Gulf Coast Studies. Texas A&M University Press, College Station, 346 pp.

FWJ at: ljudd@utpa.edu

TEXAS J. OF SCI. 59(2):127-136

MAY, 2007

EFFECTS OF TEMPERATURE AND ILLUMINATION ON
BACKGROUND MATCHING IN MEDITERRANEAN GECKOS
{HEMIDACTYLUS TURCICUS)

Frederic Zaidan III and Pamela L. Wiebusch

Department of Biology and Center for Subtropical Studies
University of Texas - Pan American
Edinburg, Texas 78539

Abstract.-The ability to change skin color is a relatively common phenomenon
in lizards whose occurrence is often related to crypsis. Field observations suggest
that Mediterranean geckos (an often ubiquitous introduced species in many
metropolitan areas of the southern United States) have the ability to lighten and
darken in response to their background. On light backgrounds, the geckos were
typically light pink whereas on dark backgrounds they were typically much darker
with a brownish hue to their skin. This study investigated the ability to background
match in this species, the main effects of temperature (20, 25, and 30°C), illumination
(total darkness and dim lighting), and their interactions on the lizard’s ability to
match their skin darkness to four levels of background darkness (black, gray, white,
and a combination of the three). Each lizard was measured in a repeated-measures
design. While temperature had little effect, illumination strongly influenced the
lizard’s ability to background contrast match. In the absence of light, 73% of the
lizards were light in color. This suggests that lighter skin pigmentation in the dark
may be the “default” setting with the melanocytes contracted. In dim lighting, the
lizard’s skin darkness closely matched background darkness in most cases (81%).
These nocturnal lizards are typically associated with human development where low
levels of illumination are often present at night. The ability of Mediterranean geckos
to accurately background match under conditions of human habitation may have
contributed to their success as a colonizing species.

Organisms may achieve crypsis by matching their own
reflectance, eolor, pattern, or a combination of the three to those of
the background (Endler 1981). The effectiveness of crypsis is
dependent on ambient light, the baekground, body coloration, and
the visual sensitivity of the viewer (Hailman 1979; Cooper &
Greenberg 1992). Color ehange in lizards is a common
morphologically or physiologically based phenomenon that is
species specific and additionally, can represent a seleetive balance
between signaling needs and the need for crypsis (Stuart-Fox et al.
2004). Morphological changes differ from physiological changes in
that morphological changes are distinguished by a gradual

128

THE TEXAS JOURNAE OF SCIENCE-VOL. 59, NO. 2, 2007

accumulation of pigment and melanophores over a period of time
(Waring 1963). Physiologieal changes tend to oeeur mueh faster,
usually within a matter of minutes, and are typieally mediated by a-
melanoeyte stimulating hormone as the darkening hormone
(Sherbrooke et al. 1994; Castrueci et al. 1997; Sherbrooke 1997;
Vazquez-Martinez et al. 2001). Physiologieal ehanges inelude
ehanges as an effect of temperature, visual responses (e.g.,
baekground eolor), or non-visual responses sueh as those invoked
by other neural or hormonal stimulation pathways.

Baekground matehing serves to minimize the amount of visual
information that a potential predator ean reeeive (Merilaita 2003).
Early studies on Gekkonid lizard coloration change show the
importanee of baekground eolor. In Atsatt (1939), eolor ehange
was measured as paling or darkening of the skin induced by the
plaeement of the white paper either around or away from the
lizard’s holding dish. Later studies showed that temperature also
had an effeet on eolor ehange in lizards beeause inereases or
deereases in temperature eould override the baekground-response to
eolor ehange (Parker 1948). At low temperatures, lizards that were
pale under a white baekground had a tendeney to turn dark, while at
high temperatures, dark lizards under a black background had a
tendency to turn pale. Color ehange in some lizards has also been
attributed to stress faetors. A study on Anolis carolinensis indieated
that eolor is most typieally affeeted by soeial aetivities. Color
fluetuation was also noted in lizards partieipating in predation
events, whether aeting as predator or prey (Greenberg 2002). Sueh
eolor fluetuations may be linked to hormones assoeiated with
physiologieal stress.

Hemidactylus turcicus (Family Gekkonidae) is an introdueed
speeies from the Mediterranean region found scattered throughout
the southern United States; the largest eontinuous range is in the
eastern and southern three-quarters of Texas and along the Gulf
Coast of Mexieo (Conant & Collins 1998). The introduetion into
Texas oeeurred slightly over 50 years ago in Brownsville, Cameron

ZAIDAN & WIEBUSCH

129

County (Conant 1955; Jadin & Coleman 2007). Throughout its
introduced range, this nocturnal species is mostly found in urban
areas on both abandoned and inhabited buildings and in cracks and
crevices of walls (urban equivalents to Middle Eastern and
Mediterranean rocky cliffs where they are naturally found; McCoy
1970). This species feeds on insects and therefore is often found
near light sources (Davis 1974). Lizards range from a translucent
shade of pink (light) to a brownish color (dark) to an intermediate
shade and individuals can change between these colors. While
mechanisms that cause color change in lizards are well known,
most studies focused on the response of diurnal species. Based on
field observations, this study investigated some factors that may be
important for color matching in a nocturnal species.

Materials and Methods

Field observations randomly chosen nights over an 18
month period (April 2004 - October 2005), 163 observations of
geckos (neonate through adult) were made from residential and
commercial areas in Hidalgo County, Texas. All studied areas lie
within 80 km of the initial site(s) of introduction in Texas and thus,
these gecko populations may represent some of the oldest in the
country. Observations are presumed to be unique, as different sites
were sampled and each lizard was likely a new observation.
Darkness of both the lizard and its background was visually scored
as light, intermediate, or dark. Visual assessment of darkness was
identical between the two authors and was further corroborated by
independent observers. A chi-square analysis was used to test if
lizard color was independent of background color. While a
quantitative measure of darkness would be preferred, the authors
feel that the simplicity and consistency of these measures were
more than adequate for a preliminary investigation.

Laboratory observations -VouxiQQn sub-adult to adult (mean
mass ± \SD = 2.05 g ± 1.48) Hemidactylus turcicus were captured
by hand from residential areas in Edinburg, Hidalgo County, Texas
(approximately 26°18’N, 98°10’W). Shortly after capture, animals

130

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

Figure 1. Typical light coloration (left) and dark (right) coloration for H. turcicus, as
seen in both the field and in the laboratory.

were taken to the laboratory in plastie buckets with ventilation
holes in the top. Lizards were identified by physical characteristics
and housed communally until tested in a 20 gallon long (75.7 L)
aquarium with newspaper substrate and numerous hiding spots.
Temperature was maintained at 24°C with a photoperiod of
12L:12D and water and commercially available crickets were
provided ad libitum. Two geckos were placed in one of four 10
gallon (37.9 L) aquaria covered with either black, gray, or white
construction paper (with the fourth consisting of all three colors).
Aquaria were placed in a Percival incubator (Perry, lA) with a
photoperiod of 12L:12D and temperature was dependent on
treatment. In the field, geckos were typically active at night-time
temperatures above 17®C, which is available for most of the year (at
the very least during early scotophase) in Deep South Texas (Starr,
Hidalgo, Cameron, and Willacy counties). Specimens were
therefore exposed to field-active temperatures. This study
investigated the main effects of temperature (20, 25, and 30°C),
illumination (total darkness and dim lighting), and their interactions
on the lizard’s ability to match their skin darkness (light,
intermediate, and dark; Figure 1) to the levels of background color

ZAIDAN & WIEBUSCH

131

Table 1. Counts of lizard shades and background shades from field observations
conducted between April 2004 - October 2005 during scotophase.

Dark

background

Intermediate

background

Light

background

Dark gecko

Intermediate gecko

Light gecko

(white = light, gray = intermediate, and black ^ dark). The gecko’s
coloration was observed once during the photophase and once
during the scotophase. Observations occurred at least two hours
after the onset of photophase or scotophase. To assess the ability to
background match, the response was scored as -2 through 2 where a
score of 0 indicated a perfect match, -1 or 1 if the gecko was one
shade lighter or darker than its immediate background, and -2 or 2
if the gecko was two shades lighter or darker. Each lizard was
rotated through all temperatures and all tank backgrounds in a
repeated-measures design. After all measurements were taken,
geckos were released at their point of capture.

Results

Early field observations indicated that the lizards change shade
in order to closely match the background shade (Table 1 and x
82.04, /^crit (0.05, 4) = 9.49, P < 0.001). Overall, the lizards correctly
matched the background 70% of the time. The lizards were most
accurate in their color matching on dark and light backgrounds
(84% and 86% correct, respectively). In contrast the lizards were
only 14% correct on intermediate backgrounds.

Because the laboratory data failed the assumption of normality,
data were analyzed with Friedman’s repeated-measures ANOVA
(Zar 1984; SAS 1985). Illumination significantly affected the
gecko’s ability to background match {P < 0.0001). The gecko’s
body coloration typically matched their background color under

132

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

dim illumination, but was lighter than their background in the
absence of light (Figure 2). No significant effects of temperature (P
= 0.2539) or interaction of temperature and lighting (P = 0.4563)
were detected.

Discussion

Field observations showed that the geckos closely matched their
background in terms of reducing their contrast. Previous studies
showed that lizards often match their reflectance or hue without an
exact color match (Norris & Lowe 1964; Gibbons & Lilywhite
1981). Any inconsistencies were due to either classification error
by the observers, differential discrimination between the lizard
versus human eye (as with the case of intermediate backgrounds),
or background matching prior to observation. Most of the obser¬
vations during this study occurred in residential areas, where
background heterogeneity is typically considerable. On several
occasions, lizards were observed to lighten or darken when moving
to new areas and the change typically took several minutes. A light
background was often adjacent to a dark one and a mismatch may
have been due to movement before observation.

Results in the laboratory setting closely reflected background
matching observed in field results. Temperature did not affect the
success of the background matching. Previous studies showed the
importance of body temperature to color change in several species
of predominantly diurnal lizards (e.g., Urosaurus ornatus,
Phrynosoma cornutum, and Sceloporus jarrovii) as a consequence
of thermoregulatory issues (Sherbrooke et al. 1994; Castrucci et al.
1997; Sherbrooke 1997). Dark coloration enhances absorption of
solar radiation and speeds an increase in body temperature (Porter
& Gates 1969). In a nocturnal lizard, such as Hemidactylus
turcicus, solar radiation is not important for thermoregulation (other
than its avoidance and during brief periods of crepuscular activity;
Carman et al. 2000) and body coloration can be used strictly for
crypsis.

ZAIDAN & WIEBUSCH

133

0.00

S -0.25 +

-0.50 --
-0.75

u -1.00

-1.25

□ dim illumination
^ complete darkness

Temperature (C)

Figure 2. Coded results for baekground matching ability for H. turcicus at three
temperatures under dim lighting and complete darkness. The color match index
represents how close the gecko darkness was to the background (0 = perfect match
and -1 = one shade lighter than background - See text). Columns represent mean
scores and error bars represent the 95% C.I.

While temperature had no effeet, illumination strongly
influeneed a lizard’s ability to match background color. In the
absence of light, 73% of the lizards were light in color. In dim
lighting, the lizard’s skin darkness closely matched the background
darkness in most cases (81%). These results suggest that: (a)
lighter skin pigmentation in the dark may be the “default” setting
with the melanocytes contracted and (b) some light is required for
the geckos to assess their immediate background. These nocturnal
lizards are typically associated with human development where low
levels of illumination are often present at night, thus light will
usually be available for the geckos to visualize their background.

Fifty years after its introduction to the southern tip of Texas,
published records of Mediterranean geckos can currently be found
for almost 90 counties (Dixon 2000; Jadin & Coleman 2007) and
many more unpublished county records likely exist. Their primary
mode of dispersal appears to be jump, rather than diffusion,
dispersal (Locey & Stone 2006). This species typically does well in

134

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

areas with vertical surfaces that support their nocturnal and

insectivorous habits (Saenz 1996; Punzo 2001). Ecological

attributes driving their success include low interspecific competi¬
tion, high survivorship, egg characteristics, and low predation
pressure (Rose & Barbour 1968; Selcer 1986). The ability to
change their color to closely match the background may allow H.
turcicus avoid the few visual predators (e.g., cats and skunks;
Locey & Stone 2006) that it would encounter during its active
period. Accurate background matching may serve to further lower
low predation pressure previously reported and contribute to its
colonization success in urban areas. This project examined
background matching on a coarse scale; further work using

quantitative measures (e.g., digital analyses) will clarify the

accuracy and precision of background matching on a fine scale and
serve to strengthen the relationship with predator avoidance.

Acknowledgements

We wish to thank J. Ortega for the photographs and for other
contributions. We would also like to thank L. Zaidan, C. Little, M.
Persans, and the Wiebusch family for assisting in various aspects of
the study. The animals were collected under Texas Parks and
Wildlife’s Scientific Collecting Permit (SPR- 1003-325) to F.
Zaidan and the studies were approved by the University of Texas -
Pan American’s Institutional Animal Care and Use Committee.
This is publication number CSS 2007-03 of the UTPA Center for
Subtropical Studies.

Literature Cited

Atsatt, S. R. 1939. Color changes as controlled by temperature and light in the
lizards of the desert regions of Southern California. Publ. University of
California at Los Angeles Biol. Sci., 1:237-276.

Carman, E. N., G. W. Ferguson, W. H. Gehrmann, C. C. Tai & M. F. Holick. 2000.
Photobiosynthetic opportunity and ability for UV-B generated vitamin D
synthesis in free-living House Geckos {Hemidactylus turcicus) and Texas Spiny
Lizards {Sceloporus olivaceus). Copeia, 2000:245-250.

ZAIDAN & WIEBUSCH

135

Castrucci, A. M. de L., W. C. Sherbrooke & N. Zucker. 1997. Regulation of
physiological color change in dorsal skin of male tree lizards, Urosaurus ornatus.
Herpetologica, 53:405-410.

Conant, R. 1955. Notes on three Texas reptiles including an addition to the fauna of
the state. Amer. Mus. Nov., 1726:1-6.

Conant, R. & J. T. Collins. 1998. A Field Guide to Reptiles and Amphibians of
Eastern and Central North America. Third Edition Expanded, Houghton Mifflin
Co., New York. xviii+616pp.

Cooper, W. E. & N. Greenberg. 1992. Reptilian coloration and behavior. Pp. 298-
422 in Biology of the Reptilia 18, Physiology E: Hormones, Brain, and Behavior
(C. Gans & D. Crews, Eds.). The University of Chicago Press, Chicago, Illinois.
xiv+564 pp.

Davis, J. R. 1974. The Mediterranean gecko, Hemidactylus turcicus in Texas. J.
HerpetoL, 8:77-80.

Endler, J. A. 1981. A predator’s view of animal color patterns. Evol. Biol., 1 1:3 19-
364.

Gibbons, J. R. H. & H. B. Lillywhite. 1981. Ecological segregation, color matching,
and speciation in lizards of the Amphibolurus decresii species complex
(Lacertilia: Agamidae). Ecology, 62:1573-1584.

Greenberg, N. 2002. Ethological aspects of stress in a model lizard, Anolis
carolinensis . Int. Comp. Biol., 42:526-40.

Hailman, J. P. 1979. Environmental light and conspicuous colors. Pp. 289-358, in
The Behavioral Significance of Color (E.H. Burtt, Jr., Ed.). Garland STPM
Press, New York, xiv+456 pp.

Jadin, R. C. & J. L. Coleman. 2007. New county records of the Mediterranean gecko
{Hemidactylus turcicus) in northeastern Texas, with comments on range
expansion. Appl. Herpetol., 4:90-94.

Locey, K. J. & P. A. Stone. 2006. Factors affecting range expansion in the
introduced Mediterranean gecko, Hemidactylus turcicus. J. Herpetol., 40:526-
530.

McCoy, C. J. 1970. Hemidactylus turcicus (Linneaus), Mediterranean gecko.
Catalogue of American Amphibians and Reptiles, 87.1-87.2.

Merilaita, S. 2003. Visual background complexity facilitates the evolution of
camouflage. Evol., 57:1248-1254.

Norris, K. S. & C. H. Lowe. 1964. An analysis of background color-matching in
amphibians and reptiles. Ecology, 45:565-580.

Parker, G. H. 1948. Animal colour changes and their neurohumours: A survey of
investigations, 1910-1943. Cambridge University Press, x+377 pp.

Porter, W. P. & D. M. Gates. 1969. Thermodynamic equilibria of animals with the
environment. Ecol. Mono., 39:227-244.

Punzo, F. 2001. The Mediterranean gecko, Hemidactylus turcicus: Life in an urban
landscape. Florida Scientist, 65:56-66.

Rose, F. L. & C. D. Barbour, 1968. Ecology and reproductive cycles of the
introduced gecko, Hemidactylus turcicus, in the Southern United States. Am.
Midi. Nat., 79:159-168.

136

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

SAS Institute. 1985. SAS Users Guide: Statistics, Version 5. SAS Institute, Inc.,
Cary, North Carolina. 956 pp.

Saenz, D. 1996. Dietary overview of Hemidactylus turcicus with possible
implications of food partitioning. J. HerpetoL, 30:461-466.

Selcer, K. W. 1986. Life history of a successful colonizer: The Mediterranean
gecko Hemidactylus turcicus in southern Texas USA. Copeia, 1986:956-962.

Sherbrooke, W. C. 1997. Physiological (rapid) change of color in homed lizards
(Phrynosoma) of arid habits. Hormonal regulation, effects of temperature, and
role in nature. Amphibia-Reptilia, 18:155-175.

Sherbrooke, W. C., A. M.de L. Castmcci & M. E. Hadley. 1994. Temperature
effects on in vitro skin darkening in the mountain spiny lizard, Sceloporus
jarrovii: A thermoregulatory adaptation? Physiol. Zook, 67:659-672.

Stuart-Fox, D. M., A. Moussalli, G. R. Johnston & I. P. F Owens. 2004. Evolution
of color variation in dragon lizards: Quantitative tests of the role of crypsis and
local adaptation. Evol., 58:1549-1559.

Vazquez-Martinez, R., J. R. Peinado, J. L. Gonzalez de Aguilar, L. Desrues, M. C.
Tonon, H. Vaudry, F. Gracia-Navarro & M. M. Malagon. 2001. Melanotrope
cell plasticity: A key mechanism for the physiological adaptation to background
color changes. Endocrinol., 142:3060-3067.

Waring, H. 1963. Color Change Mechanisms of Cold-Blooded Vertebrates.
Academic Press, New York. 266 pp.

Zar, J. H. 1984. Biostatistical Analysis, Second Edition. Prentice Hall, Englewood
Cliffs, New Jersey, xiv+718 pp.

FZ at: fzaidan@utpa.edu

TEXAS J. SCL:59(2):137-150

MAY, 2007

A COMPARISON OF MAMMALIAN DIVERSITY ACROSS
VEGETATIONAL ASSOCIATIONS OF THE ARAMBERRI REGION
OF NUEVO LEON IN NORTHEASTERN MEXICO

Jessica Valero-Padilla, Armando J. Contreras-Balderas,

Jose Ma. Torres-Ayala* and Salvador Contreras-Arquieta*

Departamento de Zoologia de Vertebrados, Laboratorio de Ornitologia and
*Departamento de Ecologia, Laboratorio de Manejo de Vida Silvestre
Universidad Autdnoma de Nuevo Leon. Apartado Postal 425
San Nicolas de los Garza, Nuevo Leon, Mexico 66450

Abstract -The mammalian fauna inhabiting nine different vegetational associa¬
tions found in two state natural protected areas (San Juan y Puentes and Trinidad y
Llano Salas) near Aramberri in southern Nuevo Leon, Mexico was examined. Prior
to this study, 32 individuals representing 13 species were collected in the zone from
August to October 2001. Monthly sampling was conducted from February 2003
through April 2004; 131 individuals of 20 species were collected and 12 individuals
of seven species were observed. In total, 25 species of mammals are reported from
this study area; these included Rodentia (15 sp.), Chiroptera (5 sp.), Lagamorpha (2
sp.). Carnivora (2 sp.) and Marsupial (1 sp.). Of the nine vegetational associations,
the Mesquital/Juniper was highest in mammalian diversity with 19 species. Three
species of rodents {Peromyscus pectoralis, Peromyscus leucopus and Neotoma
albigula) were widespread in the study area and were found in five to seven of the
vegetational associations.

Resumen.-Este estudio examina los mamiferos que habitan en nueve diferentes
tipos de asociaciones vegetales, que se encuentran en dos areas protegidas cerca de
Aramberri en el sureste de Nuevo Leon, Mexico. Antes de este estudio, se colectaron
en la zona 32 individuos (13 sp.) de agosto a octubre del 2001. Para este estudio, se
muestreo mensualmente desde febrero del 2003 hasta abril del 2004, colectandose
131 individuos (20 sp.) y observandose 12 individuos (7 sp.). En total, 25 especies de
mamiferos fueron colectadas en el area; estos incluyen Rodentia (15 sp.), Chiroptera
(5 sp.), Lagomorpha (2 sp.). Carnivora (2 sp) y Marsupial (1 sp). En cuanto a la
influencia de la vegetacion en la distribucion de los mamiferos, la mayoria de las
especies fueron colectadas en areas de MQzoydXdMJuniperus, con 19 especies. Tres
especies de roedores {Peromyscus pectoralis, Peromyscus leucopus y Neotoma
albigula) no mostraron tener preferencia de habitat, distribuyendose practicamente en
casi todos las asociaciones vegetacionales.

Under the guidelines of the “Plan Estatal de Desarrollo” the state
of Nuevo Leon in northeastern Mexico established 23 state protect¬
ed natural areas during 1997-2003 (AMAVISI 2002). Two of these

138

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

sites, San Juan y Puentes and Trinidad y Llano Salas, are loeated
near the munieipality of Aramberri in southern Nuevo Leon.

Historieally, this area of Nuevo Leon was dominated by eedar
forest and naturally occurring underground springs. During the
1970s, this area began to show the impact of increased use by
human activities. The most significant of these changes have
resulted from the clearing of the Juniper woodland for agricultural
purposes. This resulted in fragmentation of the woodland and
increased secondary vegetation. Additionally, the extraction of
groundwater for agricultural use has resulted in an increased
accumulation of organic matter (peat moss) which is subject to
spontaneous combustion. An underground fire has existed in the
area for more than 25 years.

The consequences of human impact in conjunction with the loss
of water and the presence of sulfur fumes have altered both the
diversity and distribution of the original plant and animal species.
The objective of this study was to inventory the mammalian species
present in these two state protected natural sites and determine any
existing relationships between these mammalian species and
existing vegetational associations.

Study Site

The study area is within the designated farm cooperative (Ejido)
of San Juan de Aviles and its annex Puentes is located approxi¬
mately 25 km west of the municipality of Aramberri in southern
Nuevo Leon. The Ejido de San Juan de Aviles and its annex
Puentes cover an area of 3760 ha and includes the two state
protected natural areas of San Juan y Puentes (21.7 ha, 24°09T6”N
& 100°03'37"W) with its Juniper woodland {Juniperus mono-
sperma and J. deppeana) and agricultural land and part of Trinidad
y Llano Salas (235 ha, 24°07T7”N & 100°06’27”W) which is
characterized by the halophyte vegetation of althom (Koeberlinia
spinosa) and four- wing salt bush (Atrip lex canes cens). These
protected zones are surrounded by secondary vegetation which is

VALERO-PADILLA ET AL.

139

primarily thorn brush of mesquite woodland, desert serub,
halophyte grassland and agrieultural areas.

In addition to the vegetational types, the habitats of the study
site eontain patehes of thorny eadueipholy mesquite woodland
{Prosopis glandulosa, mesquite; Opuntia imbricata, tree eholla and
Lycium berlandieri, Berlandier Wolfberry), rosetophyite desert
scrub {Larrea tridentata, creosote bush; Castela texana, amargosa;
Yucca filifera, yucca and Koeberlinia spinosa, allthom) and
halophyte grassland (Bouteloua curtipendula, sideoats grama;
Mhulenbergia monticola and navajita salina Bouteloua chasei,).

Methods and Materials

Monthly sampling was conducted from February 2003 through
April 2004. A total of 17 trap sites were monitored in nine different
vegetative associations. These were: Juniperus woodland or
Coniferus woodland (3 sites), Juniper/mesquital woodland (1 site),
Rosetophyite desert scrub (1 site), Mesquital (1 site), Mesquital/
juniper woodland (3 sites). Halophyte grassland (2 sites). Halophyte
vegetation (2 sites). Halophyte/juniper vegetation (3 sites), and
Agricultural zone (1 site). The classification of the different plant
associations was determined by the dominant plant species of the
respective areas. An attempt was made to determine the habitat
preference and distribution of each species reported from the two
study sites.

Collections of small terrestrial mammals were made utilizing
Sherman traps, Victor gopher traps and Victor mouse traps. All
traps were selectively placed near dens, runs, points of sighting, or
sites in undisturbed natural habitat. A mixture of peanut butter and
rolled oats was used as bait. The number of traps which were set
varied from 50 to 100 each night. Tomahawk traps and a .22 cali¬
ber rifle were used to collect larger species. A standard 12 m mist
net was used to collect bats. Observations and collection of data
and specimens at the study sites was conducted continuously
around the clock during the monthly sampling trips. This included

140

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

Legend
|__| study Area
• Mammals
Vegetational Association

I - 1 Juniperus Woodland

Juniperus-Mesquital Woodland
1 1 1 M Rosetophyite Desert Scrub
Mesquital

I i Mesquital- Juniperus Woodland
WB Halophyte Grassland
Halophyte Vegetation
BCBi Halophyte-Juniperus Vegetation
■■ Agricultural Zone

Figure 1. Map of the study sites showing vegetational associations and collection
localities.

road surveys, mist netting, and setting and re-visiting of the traps
(18:00 to 03:00 hrs).

Voueher specimens (skins and skulls) were prepared according
to the techniques of Hall (1981). Standard skeletal measurements
and gender were recorded for each specimen. Taxonomic identifi¬
cations were made using Hall (1981), Jimenez-Guzman et al.
(1999), Whitaker (1996), Medellin et al. (1997) and Jones &
Manning (1992). The taxonomic sequence and nomenclature
follows Ramirez-Pulido et al. (1996). All voucher specimens are
deposited in the holdings of the vertebrate collection of the
Universidad Autonoma de Nuevo Le6n (using the acronyms,
UANL-MVS) in San Nicolas de los Garza, Nuevo Leon, Mexico.

VALERO-PADILLA ET AL.

141

Results

A total of 25 mammalian species representing 10 families were
either collected or observed inhabiting the nine vegetational
associations of the two state protected natural areas of San Juan y
Puentes and Trinidad y Llano Salas in the Aramberri region of
Nuevo Leon in northeastern Mexico (Figure 1).

Mammalian Species and Vegetational Associations
Order Didelphimorphia
Family Didelphidae

Didelphis vzrgm/a^a.-Ranges from northern to southeastern
Mexico (Hall 1981). MQsqmi2i\IJumperus {\). UANL-MVS 815.

Order Chiroptera

Family Phyllostomidae

Desmodus rotundus (E. Geoffroy St.-Hilaire).-The vampire bat
occurs from eastern (north to Tamaulipas) and western (north to
Sonora) Mexico (Schmidt & Seidel 1983). MQ^Q{u.\i2i\IJuniperus
(1). UANL-MVS 797.

Family Vespertilionidae

Antrozous pallidus (Le Conte).-While the palled bat is com¬
mon, its range in Mexico is not precisely known. The reported
southern extent of this species is in the states of Jalisco and
Queretaro (Hermanson & O'Shea 1983). MQ^c^iidMJuniperus (3
captured and 2 released, 27 September 2003). UANL-MVS 796,
798 and 916.

142

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

Eptesicus fuscus (Palisot de Beauvois).-In northern Mexico, it
is most prevalent in the eastern and western Sierra Madre bordering
the arid midlands of the Mexican Plateau (Kurta & Baker 1990).
y[QS(\\x\i2i\IJuniperus (1). UANL-MVS 818.

Idionycteris phyllotis (G. M. Allen).-Mexico from Mojave
Desert scrub to fir forest, usually in the vicinity of rock outcrops
(Czaplewski 1983). MQSoyxitdMJuniperus (3). UANL-MVS 795,
904 and 917.

My Otis ciliolabrum (Merriam). -Mexico: Chihuahua, Coahuila
and Zacatecas (Holloway & Barclay 2001). MQsqmidiMJuniperus
(1). UANL-MVS 874.

Order Carnivora

Family Canidae

Canis latrans Say .-The coyote is now found between 10° N
(Costa Rica) and 70°N (Alaska). Mexico, entire country (Bekoff
1977). MQScyxitdiXIJuniperus (scat). No specimens were collected
during the course of this study. Observed on 1 March 2003.

Family Mustelidae

Taxidea taxus (Schreber).-The badger is found in the the
treeless habitats of Transition and Upper Sonoran life-zones, and is
known from arctic/alpine habitats to the lower Austral Life-zone.
Presently the badger is expanding its range eastward (Long 1973).
MQsqyiidiVJuniperus. No specimens were collected. Observed 1
individual on 24 October 2003.

Order Rodentia

Family Sciuridae

VALERO-PADILLA ET AL.

143

Sciurus alleni Nelson.-Occurs in the Mexican states of
Coahuila, Nuevo Leon, Tamaulipas and San Luis Potosi (Best
1995). Mesquital/Ji/f7z)?erw5’ (1). UANL-MVS 839.

Spermophilus variegates (Erxleben).-iS'. variegatus occurs from
the Edwards Plateu and Trans-Pecos Texas, westward through
much of New Mexico to Califoria. In Mexico, the species ranges
from Puebla, Colima, Guerrero, Mexico and Morelos northward to
the United States, although it is absent from the eastern coastal
lowlands (Oaks et al. 1987). yiQsqmXdiMJuniperus (7 and 1
observed, 24 October 2003). UANL-MVS 826, 873, 933, 940, 941,
942 and 957.

Family Geomyidae

Cratogeomys castanops (Baird).-The yellow-faced pocket
gopher occurs in Mexico; the distributional status of this species is
questionable because of the presence of two cytotypes within the
former range. It occurs south of the Rio Grande in eastern
Chihuahua and northern Zacatecas, in parts of Nuevo Le6n, and
eastward along the south side of the Rio Grande to the Gulf Coast
in Tamaulipas (Davidow-Henry et al. 1989). Halophyte vegetation
(8), Halophyte vQgQiditxonlJuniperus (2), Agricultural zone (2).
UANL-MVS 807, 813, 819, 820, 824, 899, 900, 934, 945, 946, 955
and 956.

Family Heteromyidae

Dipodomys merriami Meams.-Occurs in Mexico in the West
Gulf Coastal Plain and Central Plateau (Jimenez-Guzman 1999), in
Sierra Madre Oriental, Basin and Range, Sonoran Desert and Baja
California Peninsula regions (Hall 1981). Juniperus woodland (1),
MQsoyxitdMJuniperus (1) and Halophyte vegetation (1). UANL-
MVS 860, 906 and 922.

144

THE TEXAS JOURNAE OF SCIENCE-VOE. 59, NO. 2, 2007

Dipodomys nelsoni Merriam.-Merriam’s kangaroo rat occurs in
southern New Mexico, range extended in north central Mexico
(Best 1988). Juniper woodland (2) and Halophyte vegetation (7).
UANL-MVS 816, 821, 822, 823, 847, 853, 855, 857 and 862.

Dipodomys ordii Woodhouse.-Ord’s kangaroo rat range extends
in Mexico from Great Plains, Sierra Madre Oriental, Central
Plateau and Basin and Range regions (Hall 1981). This species
inhabits semi-arid grasslands, mixed-grasslands, scrublands
(including pinon/juniper woodlands), and sandy soils (Garrison &
Best 1990) MQsqmidiMJuniperus (1), Halophyte grassland (1),
Halophyte vegetation (2). and Agricultural zone (1). UANL-MVS
811,888, 907, 914 and 915.

Chaetodipus penicillatus (Woodhouse).-In Mexico, the desert
pocket mouse ranges from the Great Plains, Sierra Madre Oriental,
Central Plateau, Basin and Range, Sonoran Desert, and northeast
Baja California Peninsula regions (Hall 1981). Mesquital/
Juniperus (2), Desert scrub (1), Halophyte grassland (4) and
Agricultural zone (2). UANL-MVS 801, 802, 805, 806, 808, 812,
890, 902 and 905.

Order Rodentia

Family Muridae

Mus musculus Linnaeus.-The house mouse is an introduced
species and occurs throughout the American continent including all
of Mexico (Rowe & Demarest 2001). MQ^cyxiidiUJuniperus (2) and
Halophyte grassland (1). UANL-MVS 827, 828 and 846.

Neotoma albigula Hartley.-White-throated woodrats occur in
Mexico: northeastern Michoacan, and Hidalgo, Mexico (Macedo &
Mares 1988). Juniperus (12), Desert scrub (3),

Halophyte vegetation (6), Halophyte \QgQt?ii\onlJuniperus (2) and
Halophyte grassland (3). UANL-MVS 809, 814, 825, 836, 852,

VALERO-PADILLA ET AL.

145

861, 863, 869, 891, 893, 894, 908, 909, 910, 911, 923, 924, 927,
929, 935, 936, 937, 949, 951, 952 and 953,

Peromyscus difficilis (J.A. Allen)-This species occurs in
Mexico from the Sierra Madre Oriental, Central Plateau and Sierra
Madre Occidental regions (Hall 1981). yiQ^cyxit^VJumperus (1).
UANL=-MVS~829.

Peromyscus leucopus (Rafinesque).-The white-footed mouse
exhibits a range from north-central Mexico southward to the
Yucatan Peninsula (Lackey et ah 1985). Juniperus woodland (2),
MQSQyxitdi\IJumperus (2), Halophyte vegetation (2), Halophyte
grassland (4 and 1 observation) and Agricultural zone (1). UANL-
MVS 803, 817, 840, 848, 859, 867, 868, 872, 880, 901 and 931.

Peromyscus maniculatus (Wagner).-The deer mouse occurs in
Mexico in the western Sierra Madre Oriental, Basin and Range,
Central Plateau, eastern Sierra Madre Occidental, Baja California
Peninsula and Transverse Volcanic Range physiographic regions
(Hall 1981). MQsquital/ Juniperus (1), Halophyte vegetation (5),
Agricultural zone (2). UANL-MVS 832, 850, 851, 854, 856, 858,
876 and 879.

Peromyscus pectoralis Osgood.-This species inhabits the
Central Plateau and the Sierra Madre Oriental of Mexico (Schmidly
1974). Juniperus woodland (4), Desert scrub (3), Mesquital (1),
MQScyxiidiXIJuniperus (16), Halophyte grassland (14), Halophyte
\QgQtditionl Juniperus (1) and Agricultural zone (8). UANL-MVS

799, 800, 804, 810, 830, 831, 833, 834, 835, 837, 838, 841, 843,

845, 849, 864, 865, 866, 870, 871, 881, 882, 883, 884, 885, 886,

887, 889, 892, 895, 896, 897, 898, 903, 912, 913, 919, 920, 921,

925, 956, 928, 930, 932, 947, 950 and 954.

Reithrodontomys fulvescens J. A. Allen.-The fulvens harvest
mouse has a widespread geographic range centered in Mexico

146

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

(Spencer & Cameron 1982). Halophyte grassland (1). UANL-
MVS-842.

Reithrodontomys megalotis (Baird) .-The western harvest mouse
occurs in virtually all of Mexico (Webster & Jones 1982).
Halophyte grassland (1) and Agricultural zone (3). UANL-MVS-
844, 875, 877 and 878.

Order Lagomorpha

Family Leporidae

Lepus calif ornicus Gray .-Black- tailed jackrabbits range in
Mexico, from the northeast to Hidalgo and Queretaro (Best 1996).
Juniperus woodland (1 observed, 2 March 2003) Juniperus/
mesquital (1) and Halophyte vegetation (3 observed, 27 and 28
September 2003). UANL-MVS 948.

Sylvilagus audubonii (Baird).-The desert cottontail ranges from
northern to central Mexico and as far west as the Pacific coast
(Chapman & Willner 1978). Juniperus/mQ^cyxiidiX (5 and 1
observed, 26 October 2003) and Halophyte VQgQidiXAonJJuniperus (1
observed, 28 September 2003). UANL-MVS-918, 938, 939, 943
and 944.

Discussion and Conclusions

Mammalian ^'/^ec/e^'.-Jimenez-Guzman (1999) reported 41
species of mammals from the Aramberri, Nuevo Leon, Mexico
region and 33 species from the protected natural area of San Juan y
Puentes. The current study reports only 25 species from these two
protected areas, but six of these species were not reported by
Jimenez-Guzman et al. (1999). These six additional species are
Myotis ciliolabrum, Peromyscus leucopus, Peromyscus manicula-
tus, Reithrodontomys fulvescens, Reithrodontomys megalotis, and
Mus musculus. Three species of rodents {Peromyscus pectoralis,
Peromyscus leucopus, and Neotoma albigula) were widespread in

VALERO-PADILLA ET AL.

147

Table 1. Vegetational associations of the study sites and number of mammal species
collected.

Vegetation types

Area (ha)

% Total

No. species

Juniperus

136.85

Juniperus/MQsqp\i2i\

1.26

Rosetophyite Desert Scrub

145.63

Mesquital

141.29

MQsq\x\t?MJuniperus

191.91

Halophyte Grassland

145.72

Halophyte vegetation

235.05

\{2L\o^\\ytdJuniperus

90.60

Agricultural zone

1.01

Total

1089.31

100%

the study area and were found in five to seven of the vegetational
associations.

Vegetational associations vegetational associations of the
25 species of small mammals known to inhabit the Aramberri
region of southern Nuevo Leon was examined. The vegetational
type with the greatest number of species was the Mesquital/
Juniperus woodland with 19 recorded species. This was followed
by the Halophyte Grassland (8 sp.), Halophyte vegetation (7 sp.),
Agricultural zone (6 sp.), Juniperus woodland (5 sp.), Rosetofile
Desert Scrub (3 sp.), YidiXo^hyidJuniperus vegetation (3 sp.),
JuniperusMiQ^qvi\t?i\ (1 sp.) and areas of Mesquital (1 sp.).

The results of this study in the Aramberri region appear to
support the results of earlier studies conducted in the United States.
The yiQ^Q^iidXUuniperus woodland which constitutes only 18%
(Table 1) of the total area of the study site was found to provide
habitat for 76% (19 of 25) of the species of mammals reported in
this study. Sieg (1988) reported that few mammal species were
associated directly with the Juniperus woodland habitat in South
Dakota. Miller (2001) concluded that, while cedar forests do not
support a high abundance of small mammals, their presence
inhances the species diversity of adjoining grasslands by providing

148

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

a more specialized habitat. Rumble & Gobeille (1995; 2001)
reported that while many species utilize the cedar forest, the
majority of species are not associated with this type of vegetational
association.

The presence of the secondary vegetation, especially mesquite
(P. glandulosa), in the area of the cedar forest (J. monosperma and
J. deppeana) appears to have influenced the distribution of small
mammals within the study area. The secondary vegetation, cover
of this vegetational association increases food availability, nesting
materials, areas of refuge and vegetational cover. Conservation and
management practices which favor the conservation of existing
MQSC{mi2i\IJuniperus woodland would appear of primary impor¬
tance in maintaining the overall species diversity of mammals in the
San Juan y Puentes and Trinidad y Llano Salas state protected
natural areas of southern Nuevo Leon.

Acknowledgments

We would like to thank Dr. Ned Strenth for suggestions of the
draft version of this manuscript and two anonymous reviewers.

Literature Cited

AMAVISI. 2002. Plan de Manejo del Area Natural Protegida “San Juan y Puentes”.
Nuevo Leon, Mexico, 86 pp.

Bekoff, M. 1977. Canis latrans. Mammalian Species. American Society of
Mammalogists, 79:1-9

Best, T. L. 1988. Dipodomys nelsoni. Mammalian Species. American Society of
Mammalogists,326: 1 -4.

Best, T. L. 1995. Sciurus alleni. Mammalian Species. American Society of
Mammalogists, 501:1-4.

Best, T. L. 1996. Lepus californicus. Mammalian Species. American Society of
Mammalogists, 530:1-10

Chapman, J. A. & G. R. Willner. 1978. Sylvilagus audubonii. Mammalian Species.

American Society of Mammalogists, 106:1-4
Czaplewski, N. J. 1983. Idionycteris phyllotis. Mammalian Species. American
Society of Mammalogists, 208:1-4.

Davidow-Henry, B. R.; J. K. Jones, Jr. & R. R. Hollander. 1989. Cratogeomys
castanops. Mammalian Species. American Society of Mammalogists, 338:1-6.

VALERO-PADILLA ET AL.

149

Garrison, T. E. & T. L. Best. 1990. Dipodomys ordii. Mammalian Species.

American Society of Mammalogists, 353:1-10.

Hall, E. R. 1981. The mammals of North America. Vol. 1 y 2. 2a ed. John Wiley &
Sons, Inc. United States of America, 1 137 pp.

Hermanson, J. W, & T. J. O’Shea. 1983. Antrozous pallidus. Mammalian Species.

American Society of Mammalogists, 213:1-8
Holloway, G. L. & R. M. R Barclay. 2001. Myotis ciliolabrum. Mammalian
Species, 670:1-5

Jones, J. K. Jr. & R. W. Manning. 1992. Illustrated Key to Skull of Genera of North
American Land Mammals. Texas Tech University Press. United State of
America, 75 pp.

Jimenez-Guzman, A.; M. A. Zuniga-Ramos & J. A. Nino-Ramirez. 1999.
Mamiferos de Nuevo Leon, Mexico. Universidad Autonoma de Nuevo Leon,
Mexico, 178 pp.

Kurta, A. & R. H. Baker. 1990. Eptesicus fuscus. Mammalian Species. American
Society of Mammalogists, 356:1-10.

Lackey, J. A. D. G. Huckaby & B. G. Omiston. 1985. Peromyscus leucopus.

Mammalian Species. American Society of Mammalogists, 247:1-10.

Long, C. A. 1973. Taxidea taxus. Mammalian Species. American Society of
Mammalogists, 26:1-4

Macedo, R. H. & M. A. Mares. 1988. Neotoma albigula. Mammalian Species.

American Society of Mammalogists, 310:1-7.

Medellin R. A.; H. T. Arita & 6. Sanchez. 1997. Identificacion de los murcielagos
de Mexico. Asociacion Mexicana de Mastozoologia, A.C. Publicaciones
Especiales Num. 2. Mexico, 83 pp.

Miller, R. 2001. Managing Western Juniper for Wildlife. Woodland Fish and
Wildlife. MISC0286,p. 1-7.

Oaks, E. C; P. J. Young; G. L. Kirkland, Jr. & D. F. Schmidt. 1987. Spermophilus
variegatus. Mammalian Species. American Society of Mammalogists, 272:1-8.
Ramirez-Pulido, J.; A. Castro-Campillo; J. Arroyo-Cabrales, & F. A. Cervantes.
1996. Lista Taxonomica de los Mamiferos Terrestres de Mexico. Occas. Papers
Mus; Texas Tech Univ., 158:1-62. 20 de Abril de 1996.

Rowe, T. & M. Demarest, 2001. "Mus musculus" (On-line), Digital orphology.
Accessed June 27, 2006 at Web site:
http://digimorph.org/specimens/Mus_musculus/.

Rumble, M. A. & J. E. Gobeille. 1995. Wildlife Associations in Rocky Mountain
Juniper in the Northern Great Plains, South Dakota; 1994 August 8-12; Flagstaff,
AZ. Gen. Tech. Rep. RM-258. Fort Collins, CO: U.S. Department of
Agriculture, Forest Service, Rocky Mountain Forest and Range Experiment
Station, 80-90.

Rumble, M. A. & J. E. Gobeille. 2001. Small Mammals in Successional Prairie
Woodlands of the Northern Great Plains. Res. Pap. RMRS-RP-28. Fort Collins,
CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research
Station, 9 p.

Schmidly, D. J. 1974. Peromyscus pectoralis. Mammalian Species. American
Society of Mammalogists, 49: 1-3.

150

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

Schmidt, U. & M. R. Seidel. 1983. Desmodus rotundus. Mammalian Speeies.
202:1-6.

Sieg, C. H. 1988. The value of Rocky Mountain Juniper {Juniperus scopulorum)
woodlands in South Dakota as small mammal habitat. Pp. 328-332, in
Management of amphibians, reptiles, and small mammals in North America:
proceedings of the symposium; 1988 July 19-21; Flagstaff, AZ. Gen. Teeh. Rep.
RM-166. Fort Collins, CO: U.S. Department of Agriculture, Rocky Mountain
Forest and Range Experiment Station, 458 p.

Spencer, S. R. & G. N. Cameron. 1982. Reithrodontomys fulvescens. Mammalian
Species. American Society of Mammalogists, 174:1-7.

Webster, D. & J. K. Jones. 1982. Reithrodontomys megalotis. Mammalian Species.

American Society of Mammalogists, 167:1-5.

Whitaker, J. O. 1996. Field Guide to North American Mammals. National Audubon
Society. Alfred A. Knopft, New York, 937 pp.

AJC-B: arcontre@fcb.uanl.mx

TEXAS J. SCI. 59(2), MAY, 2007

151

GENERAL NOTES

AGGRESSIVE HEAD-UP DISPLAYS
IN GREAT-TAILED GRACKLES {QUISCALUS MEXICANUS)

Elissa S. Wampler and David E. Gammon

Biology Department, St. Edward's University
Austin, Texas 78704

In great-tailed grackles {Quiscalus mexicanus) the head-up
display is used by both sexes as an aggressive signal to compete for
resources such as food and mates (Johnson et al. 2000). This
display is silent and typically consists of an individual pointing its
head back while keeping its feathers sleeked against its body
(Johnson & Peer 2001). This display has been described
qualitatively (Johnson & Peer 2001), but no one has collected
quantitative data for the behavior. Both male and female great¬
tailed grackles mate with multiple partners (Johnson & Peer 2001).
These birds are sexually dimorphic, with male grackles weighing
approximately twice as much as females (Jaramillo & Burke 1999).

This study provides the first quantitative data describing head-up
displays in great-tailed grackles in relationship to sex and social
context. Both male and female grackles are known to give the
head-up display and the first objective of this study was to see if
there was a difference between the sexes in the likelihood of using
head-up displays. The second objective of this study was to
examine in which social context head-up displays were most
frequently given. Three types of social groups were defined for this
study: (1) a group composed entirely of same-sex individuals, (2) a
group in which the displaying individual is one sex but all other
individuals are of the opposite sex, and (3) a group in which the
displaying individual is one sex and the other individuals include
both males and females. For this study, a group was defined as two
or more individuals, and the average group size was four.

152

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

Great- tailed grackles were studied 10-26 Mar 2006 at five study
sites in Austin, Texas and one study site in Houston, Texas, and a
total of 75 observations were made. The study date corresponded
with the beginning of the great-tailed grackle nest-building season
(Selander 1965). All study sites were separated from each other by
2km or more. In Austin, 20 observations came from the eastern
portion of Roy Guerrero State Park, 1 6 from the western portion of
Roy Guerrero State Park, 12 from the parking lot of the HEB
grocery store on Riverside Drive, three from the parking lot of the
Albertson’s grocery store on 1-35, and three from St. Edward’s
University Campus. In Houston, 19 observations came from the
parking lot of the HEB grocery store #540 on 1-45. Birds were not
marked, but given the high numbers of grackles, it was unlikely that
many individuals were counted more than once. Most data were
gathered in the early evenings because the birds would gather in
large flocks in the evening allowing for greater social interactions.

For each observation period, an individual was chosen randomly
from a group of grackles and its behavior was observed from a
distance of approximately four to seven meters until it flew away
out of sight. This focal individual could be male or female. Each
individual studied was selected from a new group of grackles, and
was observed on average for about five minutes (range =1-10 min).
Because focal individuals were observed for different amounts of
time, this may have produced a bias in the results. No systematic
variation was noted based upon the sex of the individual or upon
the type of social group in which that individual was found;
therefore this issue was not statistically analyzed. The number and
sex of birds near the individual were recorded as well as the
presence or absence of head-up displays by the focal individual.
Only individuals within 2.5m of the focal individual were counted
when determining group composition. This distance allowed good
visual contact between the birds, so that the group’s actions and
presence could potentially influence the individual’s behavior,
resulting in head-up displays. The social group for each focal
individual was classified as ‘all same sex’, ‘all opposite sex’, or

TEXAS J. SCI. 59(2), MAY, 2007

153

‘mixed sex’, based on the sexes of the nonfocal individuals in its
group. After compiling the data, G-tests and Two Proportion Tests
were run to see if one sex was more likely to display than the other
sex, and to see if individuals were more likely to display in different
types of social groups.

More data were collected on males than females, because
females were often actively involved in nest building and did not
interact with or stay near other birds as often. Male grackles were
significantly more likely to use the head-up displays than females
(Fig. 1, Two Proportion Test, Z= 7.19, P < 0.001). Of the 53 males
observed, 72% did at least one head-up display. Of the 22 females
observed, only two individuals (9%) did a head-up display. Both of
these females used the display in groups composed entirely of other
females. In contrast, the males used the display in all three types of
social groups. Because so few females did a head-up display, the
rest of the analysis focused on male behavior.

During observation periods, focal males were typically seen
walking on the ground and did not appear to be foraging for any
great length of time; therefore the most prominent resource in the
area appeared to be the females. For the head-up display, the male
would tilt his head back, which would either illicit a head-up
response from another male, some of the birds would leave, or the
individual himself would leave. Focal males varied in the intensity
of the head-up display with some displays lasting <1 sec and others
lasting >10 sec, though most averaged two to three seconds in
length regardless of social context. The tilt of the head also varied;
in most cases the head tilted back roughly perpendicular to the
body, but in the most extreme cases the head tilted almost
completely over the bird’s back. In two instances, head-up displays
led to physical fighting between the focal male and another male,
which consisted of beating wings against each other, and grappling
with their feet. No copulations were observed during this study.

154

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

100

N=15

■o 90
a>

§ 20

I 10

I 40
o 30

0 + —

All female

All male Mixed sex

Type of Social Group

Eig. 1. Males are significantly more likely to use the head-up display when in groups of

males and females than when they are in single sex groups.

Male grackles used head-up displays in groups of all males, all
females, and mixed sex groups, but there were signifieant
differences between the three groups in the likelihood of a focal
male giving a head-up display (Fig. 2, G = 7.83, DF = 2, P =
0.020). Male grackles were significantly more likely to use a head-
up display when in mixed groups of males and females compared to
single-sex groups (Two Proportion Tests: all males vs. all females Z
= -1.43, P = 0.153; all males vs. mixed sex Z = -2.1 1 P = 0.035; all
females vs. mixed sex Z = -2.93, P = 0.003). To quantify these
differences, males used the display 45% of the time when in all
female groups, 70% of the time when in all male groups, 93% of
the time when in mixed sex groups.

In summary, great-tailed grackles used head-up displays as one
form of aggression toward other individuals. Males were much
more likely than females to use the display, and males were most

TEXAS J. SCL 59(2), MAY, 2007

155

likely to use the display in groups of mixed sex. The use of head-
up displays in this social context may allow male grackles to
achieve higher reproductive fitness, either through intersexual or
intrasexual selection. Further studies are needed to determine the
exact function of this behavior and the adaptive value of males
displaying to groups of mixed sex.

Literature Cited

Jaramillo, A. & P. Burke. 1999. New world blackbirds: the Icterids. Princeton
University Press, Princeton, NJ, 43 1 pp.

Johnson, K., E. DuVal, M. Kielt & C. R. Hughes. 2000. Male mating strategies and
the mating system of great-tailed grackles. Behav. EcoL, 1 1 : 132-141 .

Johnson, K. & B. D. Peer. 2001. Great-tailed Grackle {Quiscalus mexicanus), in The
birds of North America, No. 576 (A. Poole & F. Gill, eds.) American
Ornithologists Union, Philadelphia, PA., 28 pp.

Selander, Robert. 1965. On Mating Systems and Sexual Selection. The American
Naturalist, 99:906

ESW at: elissa.wampler@email.ucr.edu

SECOND REPORT OF THE SOUTHERN PAINTED TURTLE,
CHRYSEMYS DORSALIS (TESTUDINES: EMYDIDAE), FROM TEXAS,
WITH COMMENTS ON ITS GENETIC RELATIONSHIP TO OTHER

POPULATIONS

Chris T. McAllister, Michael R. J. Forstner and Jonathan P. Fuller

Department of Physical and Life Sciences, Chadron State College
Chadron, Nebraska 69337
Department of Biology, Texas State University
San Marcos, Texas 78666 and
625 St. Hwy. 108, Ashdown, Arkansas 71822

The southern painted turtle, Chrysemys dorsalis (Agassiz 1857)
is a medium-sized emydid turtle that ranges from western
Tennessee south through Alabama and Mississippi and westward to
eastern Arkansas, Louisiana, Oklahoma and Texas (Ernst 1971;
Iverson 1992; Conant & Collins 1998; Dixon 2002; Trauth et al.

156

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

2004). It is one of the best well-studied of all freshwater turtles
(see Ernst & Barbour 1989).

Starkey et al. (2003) recognized two evolutionary lineages
within the genus, C. dorsalis in the southern Mississippi drainage
region and Chrysemys picta from the rest of the range of the genus,
including records of the latter from Culberson and El Paso counties
in far west Texas (Strecker 1915; Carr 1952; Raun & Gehlbach
1972; Thomas 1976; Dixon 2000). Near the northeastern border of
Texas (part of the AR-LA-TX region), there are Oklahoma records
of C. dorsalis in the extreme southeastern part of the state in
McCurtain County (Webb 1970) and records from Bossier and
Caddo parishes of northwestern Louisiana (Dundee & Rossman
1989); however, this turtle has not yet been reported in
southwestern Arkansas west of the Ouachita River drainage (Trauth
et al. 2004) nor has the species been previously collected in extreme
northeastern Texas. At the western limit of its range, there appears
to be only a single genuine report of C. dorsalis from Texas (see
Dixon 2000), a specimen collected from a site near Longview on
the Sabine River, Gregg County (West Texas State University,
WTSU 6773). Other reports of C. dorsalis in Texas are based on
“sight” or “capture and release” records by wildlife biologists from
Harrison and Marion counties (Caddo Lake) and Shelby County
(Toledo Bend Reservoir) without voucher specimens available for
confirmation (see Dixon 2000). In addition, Dixon (2000, map. 48)
shows a questionable record (?) for C. dorsalis or C. picta in
Wilbarger County of northcentral Texas but it was not possible to
locate and verify the identity of this enigmatic specimen. Further¬
more, Conant & Collins (1998) placed an “x” on the map in their
field guide for C. dorsalis in central Texas, but these specimens are
known to have been introduced in Austin, Travis County, based on
information provided by W. W. Lamar, and are unfortunately, not
native (J. T. Collins pers. comm.). Therefore, this report documents
only the second authentic record of C. dorsalis from Texas with a
traditional museum voucher specimen, including some comments
on its genetic relationship to other samples of C. dorsalis.

TEXAS J. SCI. 59(2), MAY, 2007

157

Fig. 1. ASUMZ 28641, Chrysemys dorsalis from Bowie County, Texas, (a) Carapace view
showing characteristic middorsal stripe, (b) Side view, (c) Plastral view, (d) Side view
of head showing ornamentation and striping.

On 24 June 2004, an adult male C. dorsalis (carapace length =
10.5 cm) was collected by hand while it was crossing the road in
the vicinity of Barkman Creek, Bowie County, Texas (33''32.9’N,
94°6.9’W). The specimen was photographed and two months later
euthanized on 25 August 2004 with an overdose of sodium
pentobarbital (Nembutal ®). Tissues (blood, tail, leg muscle, and
testes) were taken and placed in 95% ethanol; the specimen was
subsequently fixed in 10% formalin and transferred to 70% ethanol.
Molecular methods and subsequent analyses follow Starkey et al.
(2003) inclusive of the complete suite of data included in that
publication. Genetic analyses were completed by comparing the
mtDNA D-Loop sequences from this specimen with the rangewide
samples evaluated by Starkey et al. (2003).

158

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

Texas Chrysemys

(n=^l)

C dorsalis AR and LA
(n=3 each state)

C. dorsalis IL
(n=2)

C. picta GA Clade 5
(n=3)

C. picta Rl Clade 4
(n=l)

C. picta IL and IN Clade 3
(n=3 each state)

C picta MN and Wl Clade 2
(n=2 each state)

C. picta IL Clade 1
(n=3)

Trachemys

Fig. 2. Maximum parsimony strict consensus topology from 2500 random addition full
heuristic searches. Bootstrap values from 2500 full heuristic replicates are provided
below the branches where support was greater than 50%. Descriptors for the terminal
elades follow Starkey et al. (2003). The number of individuals from each state sampled
and the state of origin within each clade are specified on the topology.

Phenotypically the specimen was typical of C. dorsalis (Fig. 1).
The carapace had a light red to orange middorsal stripe (Fig. la)
with undersides of the marginals conspicuously reddish (Fig. lb).
The plastron was yellowish and except for some widely-scattered
stains on the gular, humeral, femoral, and anal scutes, was
unmarked (Fig. Ic). The head included some yellow ornamentation
with three to four longitudinal stripes (Fig. Id). The specimen has
been deposited in the Arkansas State University Museum,
Herpetological Collection as ASUMZ 28641. As such, it represents
only the second genuine voucher of C. dorsalis from the state and

TEXAS J. SCI. 59(2), MAY, 2007

159

suggests a relatively small disjunct population occurs in the
northeastern comer of Texas.

The bidirectionally verified mtDNA sequence from this
specimen (Fig. 2) is nearly identical to Haplotype 48 (Starkey et al.
2003) and is unambiguously (all analyses bootstrap >95%)
monophyletic with other C. dorsalis samples. However, it is unique
in that it differs from other C. dorsalis haplotypes (Starkey et al.
2003) and from the complete mitochondrial sequence (GENBANK
AF069423) (Mindell et al. 1999) by a single substitution of
Cytosine for Thymine at position 16205. This unique haplotype is
unlikely to remain unique to the northeast Texas population of C.
dorsalis subsequent to future determination of genetic variation for
this taxon between the Sabine River and samples now available
from eastern and southern Louisiana.

Acknowledgments

We thank J. T. Collins, A. Cmkovic, P. Davis, J. R. Dixon, A.
Estep, R. Kazmaier, and W.W. Lamar for providing information on
locales of C. dorsalis, and S. E. Trauth (ASUMZ) for curatorial
assistance. We also thank the Texas Parks and Wildlife Department
for Scientific Collecting Permit (42-02) issued to the senior author.

Literature Cited

Carr, A. 1952. Handbook of Turtles: The Turtles of the United States, Canada, and
Baja California. Cornell Univ. Press, Ithaca, 542 pp.

Conant, R. & J. T. Collins. 1998. A Field Guide to Reptiles and Amphibians of
Eastern and Central North America. Third Edition (expanded). Houghton
Mifflin, New York, 616 pp.

Dixon, J. R. 2000. Amphibians and Reptiles of Texas. Second Edition. Texas A&M
Univ. Press, College Station, 421 pp.

Dundee, H. A. & D. A. Rossman. 1989. The Amphibians and Reptiles of Louisiana.

Louisiana St. Univ. Press, Baton Rouge, 300 pp.

Ernst. C. H. 1971. Chrysemys picta. Cat. Amer. Amph. Rept. 106:1-4.

Ernst, C. H. & R. W. Barbour. 1989. Turtles of the World. Smithsonian Inst. Press,
Washington, D. C., 313 pp.

Iverson, J. B. 1992. A revised checklist with distribution maps of the turtles of the
world. Privately printed, Richmond, 363 pp.

Mindell, D. P., M. D. Sorenson, D. E. Dimcheff, M. Hasegawa, J. C. Ast & T. Yuri.

160

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 2, 2007

1999. Interordinal relationships of birds and other reptiles based on whole
mitochondrial genomes. Syst. BioL, 48:138-152.

Raun, G. G. & F. R. Gehlbach. 1972. Amphibians and Reptiles in Texas:
Taxonomic Synopsis, Bibliography, and County Distribution Maps. Bull. 2,
Dallas Mus. Nat. Hist., Dallas, 132 pp.

Starkey, D. E., H. B. Shaffer, R. L. Burke, M. R. J. Forstner, J. B. Iverson, F. J.
Janzen, A. G. J. Rhodin & G. R. Ultsch. 2003. Molecular systematics,
phylogeography, and the effects of Pleistocene glaciation in the painted turtle
{Chrysemys picta) complex. Evolution, 57:1 19-128.

Strecker, J. K. 1915. Reptiles and amphibians of Texas. Baylor Bull., 18(4): 1-82.
Thomas, R. A. 1976. A checklist of Texas amphibians and reptiles. Texas Parks
Wildl. Dept. Tech. Ser. No., 17:1-16.

Trauth, S. E., H. W. Robison & M. V. Plummer. 2004. The Amphibians and
Reptiles of Arkansas. Univ. Arkansas Press, Fayetteville, 421 pp.

Webb, R. G. 1970. Reptiles of Oklahoma. Univ. Oklahoma Press, Norman, 370 pp.

CTM at: cmcallister@csc.com

THE TEXAS ACADEMY OF SCIENCE, 2007-2008

OFFICERS

President.

President Elect
Vice-President.

Immediate Past President:

Executive Secretary:

Corresponding Secretary:

Managing Editor:

Manuscript Editor:

Treasurer:

AAAS Council Representative:

DIRECTORS

2005 Jerry L. Cook, Sam Houston State University
Flo Oxley, Lady Bird Johnson Wildflower Center

2006 Herbert D. Grover, Hardin-Simmons University
Gary P. Garrett, Texas Parks and Wildlife Department

2007 Renard L. Thomas, Texas Southern University
Bob Murphy, Texas Parks and Wildlife Department

SECTIONAL CHAIRPERSONS

Anthropology: Roy B. Brown, Instituto Nacional de Antropologia y Historia

Botany: Alan W. Elevens, Texas Lutheran University

Cell and Molecular Biology: Jon B. Scales, Midwestern State University

Chemistry and Biochemistry: Benny E. Amey, Jr., Sam Houston State University

Computer Science: Laura J. Baker, St. Edward’s University

Conservation Ecology: Cathy Early, University of Mary Hardin Baylor

Environmental Science: Kenneth R. Summy, University of Texas-Pan American

Freshwater and Marine Sciences: Brian W. Brooks, Baylor University

Geosciences: Carol Thompson, Tarleton State University

Mathematics: Sandra Luna McCune, Stephen F. Austin State University

Physics: David Bixler, Angelo State University

Science Education: Kaycie Sullivan, Texas Tech Campus at Junction

Systematics and Evolutionary Biology: Allan W. Hook, St. Edward’s University

Terrestrial Ecology and Management: Christopher M. Ritzi, Sul Ross State University

COUNSELORS

Collegiate Academy: William J. Quinn, St. Edward's University
Junior Academy: Vince Schielack, Texas A&M University

Hudson R. DeYoe, University of Texas-Pan American

Raymond C. Mathews, Jr., Texas Water Dev. Board

William J. Quinn, St. Edward’s University

David S. Marsh, Angelo State University

Fred Stevens, Schreiner University

Cindy Contreras, Texas Parks and Wildlife Department

Ned E. Strenth, Angelo State University

Frederick B. Stangl, Jr., Midwestern State University

John A. Ward, Brooke Army Medical Center

James W. Westgate, Lamar University

SMITHSONIAN INSTITUTION LIBRARIES

3 90S8 01387 8859

THE TEXAS JOURNAL OF SCIENCE PERIODICALS

Texas Academy of Science

CMB 6252

Schreiner University

Kerrville, Texas 78028-5697

fli-l

TEXAS

THE

JOURNAL

SCIENCE

GENERAL INFORMATION

WWW . texasacademy ofscience . org

Dr. Fred Stevens, Executive Secretary
The Texas Academy of Science
CMB 6252
Schreiner University
Kerrville, Texas 78028-5697
E-mail: FStevens@schreiner.edu

THE TEXAS JOURNAL OF SCIENCE

Volume 59, No. 3 August, 2007

CONTENTS

Blanco River Symposium - Selected Papers
Symposium Coordinator: Timothy H. Bonner, Texas State University

Weathering and Water Quality in the Blanco River, a Subtropical Karst Stream.

By Michael S. Cave and Alan W. Groeger...... . . . . . . . . . 163

Spatial and Temporal Patterns in the Fish Assemblage of the Blanco River, Texas.

By Preston T. Bean, Timothy H. Bonner and Bradley M. Littrell . 179

Distribution of Cagle’s Map Turtle (Graptemys caglei) in the Blanco and
San Marcos Rivers.

By Thomas R. Simpson and Francis L. Rose . . . . . . . . . 201

Modeling Future Flows in the Blanco River Watershed under
Various Development and Rainfall Scenarios.

By Joanna C. Curran . . . . . 209

Collaborative Science and Conservation in the Blanco River Valley.

By Lacey E. Halstead and Steve Jester . . . . . . . ...233

Sponsored by:

River Systems Institute at Texas State University
www.rivers.txstate.edu

The Blanco River Symposium at Baylor University in March of 2007 and this
publication represent a joint effort by Texas State University, The Nature Conservancy of
Texas, Baylor University and the Texas Academy of Science. We wish to thank a number
of individuals for their assistance with this multidisciplinary symposium and resulting
publication. Our most sincere appreciation is extended to the Peter Way Family of
Wimberley, Texas, for serving as the primary benefactor of these efforts. Contributing
authors graciously provided their time in the preparation and publishing of their research
results. Manuscript reviewers provided helpful comments that improved the quality of
the publications. We wish to thank Frederick B. Stangl, Jr. and Ned E. Strenth of the
Texas Journal of Science for their guidance and time throughout the review and
publishing process. Lastly, we wish to extend our sincere appreciation to the Texas
Academy of Science for agreeing to host this symposium and provide a venue for the
publication of the final research results.

Andy Sansom, Director
River Systems Institute
Texas State University
San Marcos, Texas 78666

THE TEXAS JOURNAL OF SCIENCE
EDITORIAL STAFF

Managing Editor:

Ned E. Strenth, Angelo State University
Manuscript Editor:

Frederick B. Stangl, Jr., Midwestern State University
Associate Editor for Botany:

Janis K. Bush, The University of Texas at San Antonio
Associate Editor for Chemistry:

John R. Villarreal, The University of Texas-Pan American
Associate Editor for Computer Science:

Nelson Passos, Midwestern State University
Associate Editor for Environmental Science:

Thomas LaPoint, University of North Texas
Associate Editor for Geology:

Ernest L. Lundelius, University of Texas at Austin
Associate Editor for Mathematics and Statistics:

E. Donice McCune, Stephen F. Austin State University
Associate Editor for Physics:

Charles W. Myles, Texas Tech University

Manuscripts intended for publication in the Journal should be submitted in
TRIPLICATE to:

Dr. Frederick B. Stangl, Jr.

TJS Manuscript Editor
Department of Biology
Midwestern State University
Wichita Falls, Texas 76308
frederick.stangl@mwsu.edu

www.texasacademyofscience.org

AFFILIATED ORGANIZATIONS
American Association for the Advancement of Science,

TEXAS J. OF SCI. 59(3): 163- 178

AUGUST, 2007

WEATHERING AND WATER QUALITY IN
THE BLANCO RIVER,

A SUBTROPICAL KARST STREAM

Michael S. Cave and Alan W. Groeger

Department of Biology /Aquatic Station, Texas State University-San Marcos
601 University Drive, San Marcos, Texas 78666

Abstract.-The Blanco River is a karst stream that traverses the Texas Hill Country
and Balcones Fault Zone regions associated with the eastern Edwards Plateau. It is
closely connected with the Trinity Aquifer, the Balcones Fault Zone Edwards Aquifer,
and the San Marcos River, all being extremely valuable regional resources. This study
characterizes water quality in the Blanco River from the headwaters to mouth, with
particular attention given to the effects of spring and tributary inputs on concentrations of
dominant ions calcium (Ca^Q, magnesium (Mg^Q, and bicarbonate (HCO3 ). Historical
correlations between discharge and concentrations of specific ions are considered as
evidence of active diagenesis, specifically dedolomitization, in rock units of the Blanco
River. A greater concentration of dolomitic weathering products in the headwaters
region also supports this conclusion.

Karst rock, including silicates, evaporites, and carbonates have a
higher degree of solubility in natural waters than other rock types
(Gunn 1986). Carbonates such as calcite and dolomite are the most
commonly occurring of these, and are the minerals of primary interest
in this study. The dominant erosive process involved in karst
geomorphology is chemical weathering, and is largely driven by
carbonic acid, which is formed in the water or soil solution from CO2
addition. This slightly acidic solution dissolves the limestone or
dolomite crystal lattice. The ion chemistry of rivers and streams in
karst areas is normally dominated by constituents of the local
limestone in high concentrations. Over time, infiltration of water into
bedrock allows for the growth of cracks and channels that can
transport larger amounts of flow over longer distances underground.
When geological faulting and folding also occur, the dissolution of
karst rock often creates ideal conditions for the formation of aquifers.
Aquifer systems are a source of water for humans and also give rise to
springs and seeps that often support a diverse biota, including
populations of endemic organisms.

The Blanco River is a little-studied karst stream in the eastern part
of the Edwards Plateau region of central Texas. Approximately 140

164

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

km in length, it descends 250 meters in elevation and passes through
the three towns of Blanco, Wimberley, and San Marcos, Texas
(Guadalupe-Bianco River Authority 1961). The river’s 1,311 km^
drainage basin crosses two different regions of the Edwards Plateau
(or the Edwards-Trinity Aquifer), the Texas Hill Country and the
Balcones Fault Zone (Barker & Ardis 1996). Seepage from streams
draining to the east and south through the Hill Country provides
recharge to the Balcones Fault Zone Edwards Aquifer; median annual
recharge from the Blanco River between 1934 and 2004 was
estimated to be 4.40 x 10^ m^-yr'^ (35,700 acre-feet*yr‘^), or about 6%
of the total recharge to this aquifer (Edwards Aquifer Authority
2005).

The Balcones Fault Zone Edwards Aquifer supplies water to large
human populations; in 1975 it was designated the first sole-source
aquifer in Texas for the city of San Antonio (Bowles & Arsuffi 1993).
Subterranean flow in the aquifer moves mostly east-northeast and
discharges naturally at the Comal, Hueco, and San Marcos springs.
The San Marcos Springs support many endemic or range-restricted
organisms. The Blanco River supplies water to the San Marcos River
through both ground and surface water pathways. Potentiometric
surface analysis and water loss studies indicate that significant spring
recharge comes directly from the lower Blanco River (Ogden et al.
1986). The Blanco River flows into the San Marcos River 7.2 km
downstream from the springs. There the larger size of the Blanco
River watershed and the associated terrestrial inputs has a significant
effect in greatly increasing variability in physical and chemical
characteristics downstream of their confluence (Groeger et al. 1997).

Like many of the rivers and streams of central Texas, the Blanco
River is prone to fast moving, large volume flows created by the short
duration, heavy precipitation events typical to the region. Rivers in
central Texas are among the flashiest and most variable perennial
rivers in the world (Slade 1986; Poff & Ward 1989; Groeger & Bass
2005). Conversely, flows cease for extended periods in some sections
of the river, and what surface flow remains in other areas may be
entirely dependent on groundwater emerging from numerous seeps

CAVE & GROEGER

165

and springs along the river’s course as well as tributary surface flows.
The Blanco River channel is intersected by several normal faults, and
resulting fractures, bed displacements, and other structural changes
have led to a short-circuiting of surface flows. The greatest loss in
stream flow in the Blanco River occurs between 84 and 73 river km
from the mouth, where water infiltrates the bedrock in an area of
faults and highly fractured limestone created by an anticlinal flexure.
According to a previous investigation, water moves in “open fractures
or joints through the channel-entrenched carbonate” to appear as base-
flow gain downstream (Buckner & Thompson 1964). Other sections
of discontinuous surface flow are interspersed along the Blanco River
drainage.

The two major tributaries of the Blanco River are the Little Blanco
River and Cypress Creek. The Little Blanco River intersects the
Blanco River 77 km from the mouth, and it too depends on
underground flow paths. Several persistent pools that ostensibly
originate from spring inputs were observed downstream of the
monitoring site when upstream flows were nonexistent. Cypress
Creek is a 43 km stream (only the lower 22 km are perennial) which
flows through the city of Wimberley to meet the Blanco River
approximately 48 km upstream from the mouth (Bonner et al. 2002).
Cypress Creek is fed by a large limestone spring known as Jacob’s
Well (DeCook 1963). The Jacob’s Well springs stopped flowing
temporarily for the first time in recorded history in the summer of
2000, a phenomenon attributed to drought and increased development
in the Wimberley area.

The combination of above and below-ground flow paths in the
mainstem and tributary Blanco River drainages should produce
unusual relationships between discharge and the dissolved solute load.
In river systems, concentrations of dissolved ions have a strong
tendency to decrease in response to increasing discharge (Walling &
Webb 1986; Meybeck 1996). During higher flows, runoff is moved
quickly to and through the channel and has little time to pick up
solutes. Some rare cases exhibit a concentration rather than a dilution
effect in response to increasing discharge. There was a positive

166

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

correlation between discharge and specific conductance in four of the
streams contributing recharge to the Balcones Fault Zone Edwards
Aquifer (Groeger & Gustafson 1994). Two different causes for this
phenomenon in other systems have been suggested. Rains following
an extended dry period may act as a flushing mechanism for solutes
accumulated in the channel during dry periods in summer.
Alternatively, base flows in contact with a lower rock unit could have
a more dilute chemical signature than when the water table rises and
comes into contact with more soluble formations. There have been no
studies to investigate this phenomenon in subtropical karst systems,
where climate, geology, and hydrology interact in a regionally unique
way.

This study sought to characterize water quality in a subtropical
karst stream from headwaters to mouth to determine current baseline
conditions, and in conjunction with other scientists establish the
response and distribution of the riverine biological community to the
physical and chemical components of the ecosystem. This is
especially important in light of rapidly growing local human
populations and the regional stress they place upon surface and
groundwater resources.

Methods

Eight sites on the Blanco River were selected for this study with
consideration given to position, accessibility, proximity to urban
areas, and representation of different microhabitats such as riffles,
runs, or pools (Fig. 1). Site names correspond to distance upstream
from the river mouth in river km (Guadalupe-Bianco River Authority
1961). Elevations of study sites and significant landmarks taken from
the survey were compared with readings taken with a handheld GPS
unit and found to be similar. Sites were sampled on 18 dates
(monthly) from November 2003 through July 2005.

Site 127 was adjacent to private property, upstream of Blanco,
Texas and downstream of several perennial seeps that meet the river
on the property. Site 117 was located at Wayne Smith Lake, one in a
series of small overflow impoundments upstream of the Highway 281

CAVE & GROEGER

167

Distance From Mouth (km)

Figure 1. Location of study sites and map of the Blanco River drainage basin. Study
sites numbers represent their distance by km upstream from mouth. LB and CC note
confluences with the Little Blanco River and Cypress Creek, respectively. Dashed
line represents area of intermittent surface flow.

Bridge in Blanco. Sites 101 and 71 were adjacent to private property
between Blanco and Wimberley, Texas. Site 42 was located in
Wimberley approximately 6.4 km downstream from the confluence
with Cypress Creek. Sites 14 and 6 were located at road crossings in
the city of San Marcos. Additional sites on the Little Blanco River
and Cypress Creek were also selected for this study. The Little
Blanco River study site was located approximately 1 km upstream
from the confluence with the Blanco River. The Cypress Creek site
was located approximately 0.87 km upstream from the confluence
with the Blanco River.

Measurements and discrete samples were made from the bank in
areas with stream flow. Temperature, pH, dissolved oxygen, and
specific conductance were measured using a Hydrolab™ Minisonde
calibrated one day prior to each sampling event. Alkalinity was
measured by potentiometric titration to pH 4.8 using 0.02 N H2SO4
(Wetzel & Likens 2000). Turbidity was measured using a Fisher

168

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

Table 1. Median temperature and median daily temperature range (‘’C) at Sites 127, 101,
and 6 from 20 May 2004 to 2 September 2004, and at Site 101 and Valley View
Spring from 23 June 2005 to 1 1 September 2005. SD = standard deviation.

20 May 04 -2 Sept 04

23 Jun 05

-1 1 Sept 05

Site 127

Site 101

Site 6

Site 101

Valley View
Spring

Water

Temperature

27.13

27.24

28.29

29.24

22.98

SD = 2.05

SD = 1.77

SD = 1.51

SD = 1.54

SD = 0.89

Daily Range

3.30

2.99

1.65

3.32

0.17

SD= 1.85

SD = 0.70

SD = 0.56

1.13

SD = 0A\

Scientific Turbidometer. Calcium (Ca^^) and magnesium (Mg^^) were
measured by atomic absorption (American Public Health Association
1998). Soluble reactive phosphorus (SRP) was determined using the
ascorbic acid method (Murphy & Riley 1962). Nitrate-nitrogen (NO3-
N) was measured by second-derivative UV spectroscopy (Crumpton
etal. 1992).

Historical discharge and chemistry data were obtained from United
States Geological Survey (2005). Temperature was recorded every
six minutes at Sites 127, 101, and 6 from 20 May 2004 to 20
September 2004 using StowAway® XTI Temperature Loggers (Onset
Computer Corporation). The devices were placed in the best
developed channel to ensure constant immersion and shaded locations
were selected to prevent heating from direct contact with sunlight.

The section of river between Sites 101 and 71 had no visible
surface flow from 9 January 2004 to 14 March 2004. Surface flow
was observed to have ceased in the same area in the second week of
June 2005 and remained dry for the remainder of the study. A large
spring located downstream from Site 71, Valley View Spring,
appeared to be the initial source for surface flows downstream from
this area; however, additional springs with openings smaller than 1 cm
were observed nearby emerging from the bedrock of the river channel.
The orifice of Valley View Spring was much larger, approximately
one meter in width. When accessible, water from the spring was
sampled and analyzed in 2004 and 2005. Temperature loggers
described above recorded temperatures at Valley View Spring and
Site 101 every 6 minutes from 23 June 2005 to 1 1 September 2005.

CAVE & GROEGER

169

Temperature ( C)

Figure 2. Dissolved oxygen (mg/L) in relation to temperature (”C) from 2003 to 2005 in
the Blaneo River. Dashed lines indicate 80, 100, and 120% of oxygen saturation
with the atmosphere.

Results

In 2004, a relatively wet year, Blanco River flows were greater
than the quartile of historical levels from March through
December. In 2005, flows approached median historical levels after
March and continued to decrease for the remainder of the period of
study. Several high discharge events were sampled, including Site 6
on 30 June 2004, when mean daily discharge was 105 mTs (3,708
cfs). Surface flows in the Little Blanco River were typically non¬
existent in the summer at the site described above, and therefore the
data from the Little Blanco River reflects a winter sampling bias.

Mean water temperatures from the thermistors placed at Sites 127,
101, and 6 showed a clear warming trend from upstream to
downstream during the summer of 2004, corresponding to an increase
of about l.RC over 121 km, or a loss in elevation of about 180 m
(Table 1). Median daily temperature ranges and variability both

170

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

Table 2. Median values for selected variables measured in the Blanco River, Little
Blanco River, Cypress Creek, and Valley View Spring from 2003 to 2005. SD =
standard deviation; n = number of measurements.

Blanco

River

Little Blanco
River

Cypress

Creek

Valley

View

Temperature (°C)

22.98

19.38

19.98

23.08

/7= 129,5/7 = 6.36

« = 5, 5/7 = 6.38

n = 17, SD = 4.22

n^l,SD=^ 4.75

7.92

7.59

7.68

7.26

«= 129, 5Z) = 0.17

« = 5,5D = 0.15

/7= 17,5/7 = 0.14

« = 7, 5/7 = 0.16

Dissolved O2 (mg/L)

8.66

8.00

8.83

6.25

«= 124,5/7= 1.51

77 = 5,5D= 1.60

/?= 16,5D= 1.12

/7 = 7, 5/7= 1.91

Sp. Cond. (pS/cm)

439

455

522

475

f7= 129,5D = 49

n = 5, SD = 52

n= 17,5D = 38

n = 7, 5/7 = 47

Turbidity (NTU)

3.1

2.3

2.1

/7= 128,5D = 6.7

n = 6,SD = 0.5

n=n,SD = 0.1

n = 6,SD = 0.4

Alkalinity (meq/L)

3.93

4.30

4.92

4.26

«= 128, 5D = 0.54

n = 6,SD = 0.42

/7= 17,5D = 0.53

« = 6, 5D = 0.29

Calcium (meq/L)

2.32

2.56

2.95

2.72

n= 111,5D = 0.29

a7 = 4,5D = 0.17

«= 15,5D = 0.36

n =5, SD = 0.44

Magnesium (meq/L)

1.60

1.55

n= 111,5D = 0.32

/7 = 4, 5D = 0.26

n= 15,5/7 = 0.35

« = 5, 5/7 = 0.57

Ca : Mg

1.44

1.56

1.74

1.59

n= 111,5D = 0.39

n = 4,SD = 0.45

n= 15,5/7 = 0.57

« = 5, 5D = 0.49

SRP (pg/L)

4.1

2.0

3.2

5.1

n= 104,5/7 = 6.37

n = 4,SD = 7.2

n= 14,5/7 = 3.1

^7 = 4,5/7 = 0.81

NO3-N (pg/L)

316

196

209

452

«= 100,5/7 = 234

n = 4, SD = 95

/7= 13,5D= 115

77 = 4, 5D = 79

decreased moving downstream. Dissolved oxygen in the Blanco
River tended to be near saturation with the atmosphere, with 82% of
dissolved oxygen readings between 80 and 120% saturation (Fig. 2,
Table 2). Diel sampling in the river also suggested that night time
oxygen did not drop so low as to be harmful to stream organisms
(Cave 2006).

Stream turbidity was consistently low, except at Site 117 (Wayne
Smith Lake) where it was usually three to four times higher than that
of other Blanco River sites (Fig. 3a). The high mean turbidity at Site
6 reflects extremely high values recorded during storm flows on 6
June 2004 and 30 June 2004. Specific conductance did not have a
discemable trend from headwaters to mouth (Fig. 3b). Of the three
dominant ions measured in this study, specific conductance was most
strongly correlated with alkalinity {r =0.43). Alkalinity was highest
at Site 127 and decreased moving downstream (Fig. 3c). Median

CAVE & GROEGER

171

Fig. 3. Turbidity (a), specific conductance (b), and alkalinity (c) by study site (numbers
represent the distance in river km from the mouth) from 2003 to 2005 in the Blanco
River. The horizontal line and point in the boxes represent the median and mean,
respectively. The upper and lower edges of the box are the 75th and 25th percentiles.
Whiskers represent 90th and 10th percentile.

Ca:Mg ratio for the Blanco River was 1.44, but upstream of Site 71
the ratio was usually closer to 1:1 (Fig. 4a). The combined charge of
the two cations was typically equivalent to measured alkalinity. The
highest median concentration of SRP was observed at Site 71 (Fig.
4b). The greatest concentrations of NO3-N were found at Sites 117,
101, and 71 (Fig. 4c).

Compared to the Blanco River, water emerging from Valley View
Spring was low in pH and dissolved oxygen, but high in specific con-

172

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

127 117 101 71 42 32 14 6

Figure 4. Atomic ratio of calcium to magnesium (a), soluble reactive phosphorus (SRP)
(b), and NO3-N (c) by study site from 2003 to 2005 in the Blanco River. As in
Figure 3.

ductance, alkalinity, Ca^^, SRP, and NO3-N (Table 2). The therm¬
istor record reflects a time when surface flows were non-existent
between the two locations and the spring system appeared to be the
origin of flows downstream. Median water temperature of the spring
water was more than cooler than Site 101 and varied very little
over this period (Table 1).

Data collected by the USGS indicated that specific conductance,
Ca^^, and HCOs' were lowest in summer months, but Mg^^ showed
little variation from month to month (Fig. 5a). Concentrations of Ca^^
and HCOs' significantly increased with increasing discharge = 0.29

CAVE & GROEGER

173

Figure 5. Median monthly concentrations (a) of bicarbonate (HCO3), calcium (Ca^^),
magnesium (mg^^), and specific conductance from 1963 to 1997 (USGS data) (b)
same data shown in relation to discharge.

and 0.31, respectively, p<0.05), while decreased = 0.54,

p<0.05) (Fig. 5b). Specific conductance was not significantly related
to discharge in this historical data.

Discussion

Diel dissolved oxygen concentrations indicated that there was not a
problematic imbalance between photosynthesis and respiration within
the river. Greater and more variable concentrations of SRP and NO3-
N found at Sites 101 and 71 may have anthropological origins;

174

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

however, patterns in historical land use within the drainage basin were
not investigated for this study. Alternatively, spring inputs like those
from Valley View Spring enriched river water in NO3-N, though
much less so than the waters from the Balcones Fault Zone Edwards
Aquifer emerging at the Comal or San Marcos Springs (Groeger &
Gustafson 1994). Dissolved oxygen and nutrient concentrations
suggest that anthropogenic eutrophication of the river system has been
for the most part slight. With the very low surface flows common
during the typical drier summer months, point sources of nutrients
would have severe consequences for the health of this river
ecosystem.

The weathering of dolomite or magnesium-rich limestone seems to
be especially influential on solute concentrations in the Blanco River.
Springs draining dolomites or dolomite-related rocks have a Ca:Mg
ratio near unity, while that of springs draining limestone rocks are 3 to
7 times that (Shuster & White 1971). Generally, limestone containing
dolomite is much less soluble than those having a higher
concentration of calcium (Gunn 1986). Increased solubility of
dolomite at higher temperatures, however, has allowed the
development of striking karst landforms in tropical regions.
Suspended and settled precipitates observed in the Blanco River are
probably entirely calcitic, as magnesium fails to precipitate even
under extremely high saturation conditions (Land 1998). The low
median Ca:Mg ratio indicates that the most recent aggressive dolomite
solution in the drainage was probably taking place in the headwaters.
The elevated concentrations of Mg^^ upstream appeared to be diluted
downstream by waters draining rock units richer in calcite.

A portion of the river’s water is supplied by two major tributaries,
which are both, for the most part, similar in origin and chemical
character to the Blanco River. The intermittent Little Blanco River
tributary seemed to have little effect on Blanco River chemistry
during the study, but joins the Blanco River in the region in which
both were dry under low flow conditions. Therefore the upstream
Little Blanco River surface flow, which is usually present, is probably
entering the same or a parallel underground flow pathway through

CAVE & GROEGER

175

which the upstream Blanco River water has moved. Water from
Cypress Creek, which is often quantitatively important during low
flow periods (Buckner & Thompson 1964), had an observable effect
on the Blanco River. Elevated pH, specific conductance, and Ca^^
recorded at the site downstream of the confluence (Site 42) probably
resulted from solute-rich inflows from spring-fed Cypress Creek. The
increase in solute concentrations could also have been influenced by
spring outflows in the Blanco River channel upstream of the
confluence. A significant stream flow gain was traced to springs in an
area between 18.3 and 19.3 km upstream from the mouth of Cypress
Creek (Buckner & Thompson 1964).

Groundwater from springs like Valley View Spring may be a
significant driver for temperature and dissolved ion dynamics in the
Blanco River. Valley View Spring water was cooler than Blanco
River surface water in the summer and warmer in the winter,
indicating that this water had a much slower underground transit time
than above ground. Groundwater was found to be an important source
and sink for thermal energy for a Pennsylvania karst stream,
depending on season, and stream temperatures were strongly related
to surface and groundwater interactions (O’ Driscoll & DeWalle
2006). Chemically, the spring water was enriched in Ca^^ and
alkalinity relative to Blanco River surface water and was lower in pH.
The lower pH in the spring water suggests that CO2 introduction is
occurring more rapidly than the process of calcite or dolomite
dissolution (Thrailkill 1972).

Texas surface waters show a strong trend of decreasing specific
conductance from west to east corresponding to increased rainfall
(Groeger & Ground 1994). However, in the Edwards Plateau region,
those systems farthest to the east exhibited higher specific
conductance than those to the west (Groeger & Gustafson 1994). The
drainages in this region clearly possess hydrological and
geomorphological features that exhibit a unique chemical response to
changing hydrological conditions. Storm flow data suggest a
threshold for the positive relationships between discharge and
concentrations of Ca^^ and HCOs'. Samples from Site 6 during high
discharge events on 10 June 2004 and 30 June 2004 do not reflect a

176

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

proportional increase in dissolved solute concentration, but rather they
were more dilute with respect to median values from 2003 to 2005.
However, samples associated with storm flows were obtained on the
falling limb of the hydrograph, and it was not determined whether this
was the late stage of a ‘‘flushing” effect.

Summer decreases in concentrations of Ca^^ and HCOs' probably
result from a combination of lower discharge and higher daily
temperatures. Solubility of calcium carbonate decreases with
increasing temperature (Drever 1997). Historically, Ca^^ and HCOs'
in the Blanco River decreased with increasing temperature (Fig. 5 A, r
= 0.43 and 0.42, respectively, p<0.05). There was no relationship
between Mg^^ concentrations and temperature in the historical data.

The process of active diagenesis in Blanco River rock units could
explain both the relationships between discharge and dominant ion
concentrations as well as the greater abundance of dolomitic
weathering products in the headwaters. In the late Miocene period,
faulting along the Balcones fault zone raised the Edwards Plateau in
the north and west relative to sea level and enabled the formation of a
circulating freshwater aquifer (Ellis 1986). Since that time, rocks in
the aquifer underwent several major near- surface diagenetic changes,
including extensive dedolomitization. In this process, freshwater
flushing replaces gypsum and magnesium in dolomitic rocks with
calcite. The resulting dedolomite can be more soluble than the
original dolomite (Evamy 1967). Isotopic ratios suggest dedolo¬
mitization continues in the present-day (Ellis 1986). High concen¬
trations of sulfate, possibly resulting from dissolved gypsum, have
been recorded in the upper Blanco River (Guadalupe-Bianco River
Authority 2003). Sulfate concentrations, like Mg^^, exhibited a
negative relationship with flow. Increased discharge could allow
water in subterranean flow paths to reach more soluble, magnesium-
poor calcites, creating the concentration effect observed for Ca^"^ and
HCOa' in the historical record (Fig. 5a). During storm flows, the
amount of water flowing through the system overwhelms the
concentration effect and dissolved solutes become diluted.

The streams recharging the Balcones Fault Zone Edwards aquifer
are similar in chemistry (Groeger & Gustafson 1994). Like the

CAVE & GROEGER

177

Blanco River, spatial and temporal patterns in the ion content of each
stream are presumed to be driven by the physical and chemical
weathering processes occurring in their underlying geological forms.
Past and present data suggest that these streams would not exhibit
higher salinities in a drier climate, thanks in large part to unique
solution mechanics affecting the local karst and the tight connectivity
between surface and groundwater flow paths. Endemic aquatic
organisms depend on water from springs and seeps to maintain base
flows as well as the consistent thermal and chemical conditions to
which they have adapted. The perpetuity of these valuable
ecosystems and their associated biotic assemblages will depend on
human diligence in preserving the supply of environmental
groundwater flows.

Literature Cited

American Public Health Association. 1998. Standard methods for the examination of water
and wastewater, 20**’ edition. American Public Health Association, Washington D.C.,

1268 pp.

Barker, R. A. & A. F. Ardis. 1996. Hydrogeologic framework of the Edwards-Trinity
Aquifer system, west-central Texas. U.S.G.S Prof Paper 1421-B. Washington D.C., 61

pp.

Bonner, J. S., F. J. Kelly, M. Beaman, & R. Wilkinson. 2002. Impairment verification
monitoring - Vol. 1: Physical and chemical components Segment 1815 Cypress Creek.
Report to the Texas Commission on Environmental Quality, 211 pp.

Bowles, D. & T. Arsuffi. 1993. Karst aquatic ecosystems of the Edwards Plateau region of
central Texas, USA: a consideration of their importance, threats to their existence, and
efforts for their conservation. Aquat. Conserv., 3:317-329.

Buckner, H. D. & G. L. Thompson. 1964. Base flow study: Blanco River, Texas February-
March 1963. United States Department of the Interior, Geological Survey Water
Resources Division, Open File No. 70- June, 1964, 13 pp.

Cave, M. S. 2006. Effects of surface and groundwater interactions on the solution chemistry
of a subtropical karst stream. Unpublished M. S. thesis, Texas State Univ., San Marcos,
47 pp.

Crumpton, W. G., T. Isenhart, & P. Mitchell. 1992. Nitrate and organic N analyses with
second-derivative spectroscopy. Limnol.Oceanogr., 37:907-913.

DeCook, K. J. 1963. Geology and ground-water resources of Hays County, Texas. United
States Government Printing Office, Washington D.C., 23 pp.

Drever, J. 1997. The geochemistry of natural waters: Surface and groundwater
environments. Prentice Hall, Englewood Cliffs, New Jersey, 467 pp.

Edwards Aquifer Authority. 2005. Edwards Aquifer Authority Hydrologic Data Report for
2004, SanAntonio, TX, 129 pp.

Ellis, P. M. 1986. Post-miocene carbonate diagenesis of the Lower Cretaceous Edwards
Group in the Balcones fault zone area, south-central Texas. Pp. 101-114, in The
Balcones Escarpment, Central Texas (Abbott, P.L. & C.M. Woodruff, eds). Geological
Society of America, pp. 200.

178

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

Evamy, B. D. 1967. Dedolomitization and the development of rhombohedral pores in
limestones. J. Sediment. Petrol., 37:1204-1215.

Groeger, A. W. & D. A. Bass. 2005. Empirical predictions of water temperatures in a
subtropical reservoir. Arch. HydrobioL, 162:267-285.

Groeger, A. W. & T. A. Ground. 1994. Salinity and ionic composition of Texas reservoirs.
Arch. Hydrobiol. Ergebn. LimnoL, 40:27-33.

Groeger, A. W. & J. J. Gustafson. 1994. Chemical composition and variability of the waters
of the Edwards Plateau, central Texas. Pp. 39-46, in Second international conference on
ground water ecology. (J.A. Stanford, & H. M. Valett, eds.), American Water Resources
Association, Herndon, VA, 390 pp.

Groeger, A. W., P. F. Brown, T. E. Tietjen, & T. C. Kelsey. 1997. Water quality of the San
Marcos River. Texas J. Sci., 49(4):279-294.

Guadalupe-Bianco River Authority. 1961. Appendix 1, Exhibit 5: Profile - Guadalupe
Tributaries - Blanco River. In: Report on supplement to the initial plan of development.

Guadalupe-Bianco River Authority. 2003. Investigation of elevated sulfate concentrations in
the Upper Blanco River: Report. Guadalupe-Bianco River Authority. Seguin, TX.

Gunn, J. 1986. Solute processes and karst landforms. Pp. 363-437, in Solute processes (S. T.
Trudgill, ed.) John Wiley and Sons Ltd., London, 386 pp.

Land, L. S. 1998. Failure to precipitate dolomite at 25 degrees C from dilute solution despite
1000-fold oversaturation after 32 years. Aquat. Geochem., 4:361-368.

Meybeck, M. 1996. River water quality: Global ranges, time and space variabilities,
proposal for some redefinitions. Proc. Intemat. Assoc. Theor. Appl. LimnoL, 26: 81-96.

Murphy, J. & J. Riley. 1962. A modified single solution for the determination of phosphate
in natural waters. Anal. Chim. Acta, 27: 31-36.

O’Driscoll, M. A. & D. R. DeWalle. 2006. Stream-air temperature relations to classify
stream-ground water interactions in a karst setting, central Pennsylvania, USA. J.
Hydrol., 329:140-153.

Ogden, A., R. Quick, & S. Rothermel. 1986. Hydrochemistry of the Comal, Hueco, and San
Marcos Springs, Edwards Aquifer, Texas. Pp. 115-130, in The Balcones Escarpment,
Central Texas (P. L. Abbott, P. L. & C. M. Woodruff, eds.). Geological Society of
America, 200 pp.

Poff, N. L. & J. V. Ward. 1989. Implications of streamflow variability and predictability for
lotic community structure: a regional analysis of streamflow patterns. Can. J. Fish.
Aquat. Sci., 46:1805-1881.

Shuster, E. T. & W. B. White. 1971. Seasonal fluctuations in the chemistry of limestone
springs: a possible means for characterizing carbonate aquifers. J. Hydrol., 14:93-128.

Slade, R. M. 1986. Large rainstorms along the Balcones Escarpment in central Texas. Pp.
77-90, in The Balcones Escarpment, Central Texas (P. L. Abbott, & C. M. Woodruff,
eds). Geological Society of America, 200 pp.

Thrailkill, J. 1972. Carbonate chemistry of aquifer and stream water in Kentucky. J.
Hydrol., 16:93-104.

United States Geological Survey. 2005. USGS 08171000 Blanco Rv at Wimberley, TX.
http://waterdata.usgs.gov/tx/nwis/nwisman? Retrieved on 2005-10-30 16:41:33 EST.

Walling, D. E. & B. W. Webb. 1986. Solutes in river systems. Pp. 251-327 in Solute
processes (S. T. Trudgill, ed.), John Wiley and Sons Ltd., London, 386 pp.

Wetzel, R. G. & G. E. Likens. 2000. Limnological Analyses. Springer, New York, 360 pp.

AWG at: agl l@txstate.edu

TEXAS J. OF SCI. 59(3): 179-200

AUGUST, 2007

SPATIAL AND TEMPORAL PATTERNS IN
THE FISH ASSEMBLAGE OF THE BLANCO RIVER, TEXAS

Preston T. Bean, Timothy H. Bonner
and Bradley M. Littrell

Department of Biology/Aquatic Station
Texas State University-San Marcos, 601 University Drive
San Marcos, Texas 78666

Abstract.-Spatial and temporal patterns in occurrence, abundance, and habitat
associations of the Blanco River drainage fish assemblage were examined among ten
sites that were sampled quarterly for two years. Cyprinids comprised 78% of the overall
assemblage, with Cyprinella venusta (41%), Pimephales vigilax (14%), and Notropis
amabilis (11%) being the most abundant species. Variation in the fish assemblage was
examined using canonical correspondence analysis. Physical habitat parameters
explained 15.3%, followed by site (11.2%), and season (2.3%). The impact of low-head
dams on the fish assemblage was also assessed. Low-head dam impoundment
assemblage was markedly different from riverine mainstem sites (Analysis of
Similarities, P < 0.01) in that the impoundment assemblage was dominated by more
lentic species and generally lacked species normally associated with higher velocity runs
and riffles.

Identification of patterns in species diversity and abundance and
their causal mechanisms have received much attention (Shmida &
Wilson 1985; Brown & Maurer 1989). The causal mechanisms
generally are subdivided into abiotic and biotic factors, and are
evaluated on recent and localized scales (Brown & Maurer 1989;
Matthews 1998). Abiotic factors include both physical and chemical
characteristics of a stream (i.e., depth, current velocity, substrate,
temperature, pH, dissolved oxygen, and turbidity) and can affect
assemblages based on the autecology of species (Whiteside & McNatt
1972; Matthews 1998). Among the many abiotic factors, gradients in
current velocity, depth, and substrate often most strongly associate
with variation in fish assemblages at a local scale (Gorman & Karr
1978; Schlosser 1982; Cantu & Winemiller 1997; Walters et al. 2003;
Williams et al. 2005). Biotic factors affecting fish assemblages
include: intra- and interspecific competition, food availability, and
predation (Matthews 1998). Understanding which factors are most
strongly associated with the distributional patterns of stream fishes
can reduce the error in predicting how fish assemblages might change

180

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

as habitats are impacted by anthropogenic effects (Harding et al.
1998).

Although the southern United States has a highly diverse fish fauna
(Burr & Mayden 1992, Warren et al. 1997), 28% are listed as extinct,
endangered, threatened, or of special concern (Warren et al. 2000).
Williams et al. (1989) list five factors contributing to the demise of
North American fishes: habitat degradation, over-exploitation,
disease, natural or anthropogenic-induced biotic factors, and restricted
range. Warren et al. (2000), however, attribute the decline of native
fishes of the southern United States primarily to habitat degradation.

Anthropogenic disturbance in the form of increased sediment and
nutrient loads, introduced species, and altered hydrologic regimes
associated with dams, is among the greatest threats to the freshwater
fauna of the United States (Richter et al. 1997), and is cited as the
reason for species declines across the country (Warren et al. 2000).
Impoundment of streams reduces the connectivity of upstream and
downstream segments (Edwards 1978) and among streams within a
drainage (Herbert & Gelwick 2003), decreases the discharge (Bonner
& Wilde 2000) and magnitude of floods (Adams 1985) downstream
from impoundments, and creates a more lentic habitat within the
impounded segment (Taylor et al. 2001). Collectively, these
alterations adversely affect fish assemblages. Reduced stream
connectivity restricts movement of fishes, resulting in reduced
upstream diversity, extirpation of obligate riverine species (Winston et
al. 1991; Porto et al. 1999), and dominance of the assemblage by
habitat generalists (Winston et al. 1991; Taylor et al. 2001). When
variable upstream reaches experience harsh conditions that cause a
local extirpation of all fish from a stream reach, these reaches are
subsequently repopulated by species surviving in stable, downstream
habitats (Whiteside & McNatt 1972). Impoundments serve as a stable
source from which upstream habitats are opportunistically colonized
and are generally dominated by generalist species (Herbert & Gelwick
2003). With obligate riverine species being absent from reservoirs,
even temporary cessation of flow upstream from an impoundment
might lead to the permanent extirpation of fluvial specialists and an
assemblage dominated by habitat generalists. Downstream from
dams, changes in habitat caused by scouring of substrate (Gillette et

BEAN, BONNER & LITTRELL

181

al. 2005) and reduced peak discharge (Bonner & Wilde 2000)
contribute to changes in fish assemblages.

The purpose of this study was to identify factors important in
structuring the Blanco River fish assemblage and determine the
effects of low-head dams within the watershed. The Blanco River is a
stream system typical of the Texas hill country and Edwards Plateau
characterized by low turbidity and high dissolved solids.
Additionally, these streams possess many endemic taxa, about which,
little is known. These stream systems face several threats (Bowles &
Arsuffi 1993) including low-head dams as Texas leads the nation with
over 6,000 dams constructed in its waters (Shuman 1995). The effects
of low-head dams on the fish assemblages of streams of the Texas Hill
Country are not known. Description of the current fish assemblage
and identification of factors structuring the assemblage will allow for
determination of future changes in the Blanco River fish assemblage
and prediction of impacts of anthropogenic disturbance within the
watershed on regional and drainage endemic species (e.g., Dionda
nigrotaeniata, Macrhybopsis marconis, and Micropterus treculii) as
well as the overall assemblage. Specifically, the objectives were to
determine current habitat and fish assemblage structure and identify
habitat associations, longitudinal and seasonal patterns, and effects of
low-head dams on the Blanco River fish assemblage.

Methods

The Blanco River drains an area of 1,067 km^ (USGS 2003) in
Kendall, Comal, Blanco, and Hays counties, Texas, before its
confluence with the San Marcos River (Fig. 1). Little Blanco River
and Cypress Creek are the two largest tributaries of the Blanco River.
Both tributaries are spring-fed although baseflow in the Little Blanco
River is subterranean about 5 km before reaching the Blanco River.

Ten sites in the Blanco River watershed were sampled quarterly
from October 2003 through July 2005. Eight sites were located on the
mainstem with two upper (sites 1 and 2), two middle (sites 3 and 4),
and four lower reach sites (sites 5, 6, 7, and 8). Two sites were
established on major tributaries with one on the Little Blanco River
(Site 9) and one on Cypress Creek (Site 10).

182

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

Figure 1: Map of study sites within Blanco River watershed sampled quarterly from

October 2003 through July 2005.

At each site, fish were collected from available geomorphic units
{i.e., runs, riffles, pools, backwaters, reservoirs, and plunge pools;
Arend 1999) by a combination of seining (9.5 mm mesh), backpack
electrofishing (Smith-Root Model 12-B POW), and experimental gill
nets (three nets set for two hours). Seines were used at all sites,
backpack electrofishing was used in areas not conducive for seining
(i.e., around cover, large woody debris, boulders, and shallow riffles)
and gill nets were used at Site 2 (Reservoir Site) in deepwater (>2 m)
habitats. Fish were collected from each geomorphic unit until fish
were depleted from the geomorphic unit (only a few individuals were
captured) and no new species was collected (Williams & Bonner
2006). Fish from each geomorphic unit were isolated in buckets until
sampling was completed in all geomorphic units. Fish were identified
to species, measured to the nearest millimeter total length (up to 30
specimens per species per site), and released or retained as voucher
specimens. Voucher specimens were anesthetized with a lethal dose
of tricaine methanesulfonate and preserved in 10% formalin.

Habitat parameters recorded include geomorphic unit type, length,
stream width, percent substrate (silt, sand, gravel, cobble, boulder, and
bedrock), percent woody debris, percent vegetation, percent detritus,
mean current velocity (m/s), maximum current velocity (m/s), mean
depth (m), maximum depth (m), temperature (°C), pH, conductivity
(pS/cm), dissolved oxygen (mg/L), and turbidity (NTU). Geomorphic

BEAN, BONNER & LITTRELL

183

unit length and width were measured to the nearest meter. Percent
substrate^ woody debris, vegetation, and detritus were visually
estimated for each geomorphic unit (Williams et al. 2005). Depth was
recorded to the nearest 0.01 m and current velocity was measured
using a Marsh-McBimey FLOW-MATE™ model 2000 flow meter.
Temperature and chemical parameters were measured using YSI-
Model 85 and YSI-Model 650 water quality instruments. Site
estimates of physical and chemical data were calculated by weighted
averaging by geomorphic unit area. Stream discharge was obtained
from USGS gauging stations No. 08171000 (Wimberley, Texas) and
No. 08171300 (Kyle, Texas).

Principal components analysis (PCA) was performed using site
means of physical habitat data. Qualitative data (i.e., geomorphic
units) were represented with dummy variables whereas quantitative
data were z-score transformed (Krebs, 1999). The resulting loadings
and plots were used to describe habitat present at each site. Fish
abundance and habitat data were analyzed using canonical
correspondence analysis (CCA; Canoco 4.5, ter Braak 1986) to
determine habitat associations as well as seasonal, site, and habitat
effects in structuring the Blanco River fish assemblage. A variance
partitioning method (Borcard et al. 1992) was used to determine pure
site, season, and habitat effects as well as shared (two- and three-way)
effects by producing a reduced CCA model for each effect with the
additional two effects as covariates. Species richness, Shanon-Wiener
diversity indices, and Pielou’s evenness indices were calculated in
PRIMER (version 5; Primer-E, Ltd., Plymouth, United Kingdom) for
each site per each quarter. Bray-Curtis similarity indices were
calculated for species abundance data pooled by season for each site.
Species abundance data were standardized as relative abundances
because sampling effort (i.e., area sampled) was not equal among
sites. The resulting similarity matrix was used in analysis of
similarities (ANOSIM; Clarke & Green 1988; Clarke 1993) to test for
differences in fish assemblage structure among sites within, adjacent
to, and distant from impoundments. Sites adjacent to impoundments
were defined as those within 1 km of an impoundment whereas sites
distant from impoundments were greater than 1 km from an
impoundment. Distance to impoundment was determined along the

184

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

thalweg by examination of aerial photographs and topographical
maps. Determination of which species were contributing the greatest
amount to the dissimilarity between the impoundment and the other
categories was accomplished using the SIMPER function in PRIMER.

Results

The Blanco River, Little Blanco River, and Cypress Creek
generally were shallow to moderate depth wadeable streams with
substrate dominated by bedrock with some coarse gravel (Table 1).
Sites ranged in length from 43 to 325 m depending on heterogeneity
of available habitats and accessibility. The tributaries. Little Blanco
River and Cypress Creek, were deeper, had slower current velocities,
greater percentages of gravel substrate, and more aquatic vegetation
and detritus than mainstem sites. Mainstem sites were wider and
shallower with swifter current velocities and greater percentages of
bedrock substrate. Sand and boulder substrates were uncommon
across all sites but were both highest at Site 5. Among mainstem
sites, the Reservoir Site (Site 2) and Site 8 had the greatest amount of
vegetation (filamentous algae and emergent macrophytes: 22%)
whereas no vegetation was present at Site 5. During the duration of
the study, median discharge was 4.25 m^/s between sites 4 and 5
(USGS Station No. 08171000) and 3.96 mVs between sites 6 and 7
(USGS Station No. 08171300). Stream discharge was lowest in late
summer and fall and increased sharply in the spring in both years of
the study (Fig. 2). Across sites and dates, mean temperature was
20°C, pH was 9.18, conductivity was 441 pS/cm, dissolved oxygen
was 8.77 mg/L, and turbidity was 2.9 NTU. Seasonal and diel water
quality and geochemistry measurements were taken concurrently with
this study and reported by Cave (2006).

Principal component axes I and II explained 37% of the variation
in habitat among sites within the Blanco River watershed. PC I
represented a substrate gradient whereas PC II represented a velocity,
depth, and substrate gradient (Fig. 3). Strongest positive loadings for
PC I were gravel (0.52), woody debris (0.34), and vegetation (0.32);
strongest negative loadings were bedrock (-0.52), boulder (-0.16), and
sand (-0.16). Strongest positive loadings for PCII were silt (0.56),

BEAN, BONNER & LITTRELL

On bo

K) V/i

—

^ VO

K) —

— O

tvj —

so — —

— (-rt

>— O

— ^ ^ ^

— u>

U) —

I— K)

— O

— o

K> — ^

—

— UJ

185

Table 1. Mean (± SD) physical habitat parameters across sampling dates for eight sites on the Blanco River, one site on the Little
Blanco River, and one site on Cypress Creek sampled between October 2003 and July 2005.

_ (Reservoir) _ Main Stem _ Little Blanco Cypress Creek

186

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

Figure 2: Median monthly discharge in the Blanco River (USGS Station No. 08171300)
from October 2003 through July 2005.

depth (0.38), and vegetation (0.36); strongest negative loadings were
current velocity (-0.58), bedrock (-0.18), and gravel (-0.16). Site 1
and the Reservoir Site had higher percentages of bedrock and silt
substrate. The Reservoir Site differed from Site 1 by having greater
vegetation and depth. Site 4 had a high percentage of bedrock and
gravel and the lowest percentage of silt. Sites 5 and 6 had greater
percentages of bedrock and boulder substrates and higher current
velocities whereas sites 3 and 8 and Little Blanco River had greater
percentages of gravel substrate with sites 3 and 8 having greater
percentages of cobble and sites 8 and Little Blanco River having
greater amounts of woody debris. Cypress Creek had a greater
percentage of bedrock and cobble substrate and detritus.

A total of 29,265 fishes representing 10 families and 33 species
was collected from October 2003 through July 2005 within the Blanco
River watershed (Table 2). Overall fish abundance was highest at Site

BEAN, BONNER & LITTRELL

187

Silt (0.56)
Depth (0.38)
Vegetation (0.36)

Velocity (-0.58)
Bedrock (-0.18)
Gravel (-0.16)

Bedrock (-0.52) pc r travel (0.52)

Sand (-0. 1 6) Woody debris (0.34)

Silt (-0.11) • Cobble (0.31)

Figure 3: PCA habitat plots of PC axes I and II for the Blanco River, Little Blanco
River, and Cypress Creek. Upper graph represents all site and date combinations. Lower
graphs represent each site at eight sampling dates.

Table 2. Relative abundances of fishes collected from the Blanco River, Little Blanco River, and Cypress Creek across all
sampling dates.

188

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

rt fS| — 00

O O O O

; tN «N

o d d <N

<N , \0

d d d

d ' -rj-' cs

O 1^0 (N ;0O —

d !I;;'dd ddddd

-.s - .

^ ^ ^ o a <=> OOvOo*^ ~<=>

to CO g <N ;;;
O O

(N m

tn o K

I ^ ^ I 'R “

o O O O (N w-i o

d d

— d ® CO

<N vq
CO d

d d

lO >

o I

d “=> ®

CO ^ ^ >0

d d d

, '>q O , >q

2 — ’ d d d d

IT) I I 'O >0 ^ I ^

d ' ’ d d ‘ ' d

CO <N , >0

d d d

dd ddd ddd

fN CO

C'l o

^ ,rs ,oSS®dg

o -< ‘Nod'^'^d

VO Ov OOO-^vnO*^^
O f^OO’rf■r^O

^ w O ■•j
*>^ **' ^ U

;§ :§ tq 0., a. O

BEAN, BONNER & LITTRELL

189

2 (N= 6,586) and Site 8 (AA- 4,319) and lowest at Site 4 (TV = 1,213)
and Cypress Creek (TV = 1,273). Most abundant families were
Cyprinidae (78%), Centrarchidae (10%), and Poeciliidae (8%).
Lepisosteidae, Catostomidae, Characidae, Ictaluridae, Fundulidae,
Percidae, and Cichlidae each comprised less than 2% of the overall
assemblage. Cyprinella venusta (41%), Pimephales vigilax (14%),
Notropis amabilis (11%), Gambusia affinis (8%), and Notropis
volucellus (5%) were the most abundant species. Cyprinidae (TV of
species =11) and Centrarchidae (TV of species =10) were the most
species rich families. Species richness ranged from 3 to 15 among
samples with the lowest richness occurring at Site 4 in January 2004
and the highest richness occurring at Site 8 in October 2003 and at
Site 7 in July 2005. Mean Shannon Diversity and Pielou’s evenness
were highest at the Little Blanco River and Cypress Creek and lowest
at Site 5 (Table 3). Highest individual sample diversities occurred at
the Little Blanco River in July 2005 (2.37) and July 2004 (2.09).
Lowest individual sample diversities occurred at Site 1 (0.33, July
2004) and Site 4 (0.54, January 2005).

Cyprinids generally were more abundant in the mainstem and
Cypress Creek, systems with more persistent flows, whereas
centrarchids and poeciliids generally were more abundant in Little
Blanco River, a stream with intermittent flows near its confluence
with the Blanco River mainstem. Cyprinidae (81%) was the most
abundant family in the Blanco River, followed by Poeciliidae (8%)
and Centrarchidae (7%). Likewise, Cyprinidae (69%) was the most
abundant family in Cypress Creek, followed by Centrarchidae (20%),
Poeciliidae (4%), Percidae (4%), and Cichlidae (3%). In the Little
Blanco River where pool habitats were common, Centrarchidae (42%)
was the most abundant family, followed by Cyprinidae (38%),
Poeciliidae (14%), Percidae (4%), and Catostomidae (2%).

Among fishes with a relatively small geographic range, Dionda
nigrotaeniata was only present in the tributaries with its greatest
abundance occurring in the Little Blanco River (1.8%). Three
individuals of Macrhybopsis marconis were collected at Site 8.
Notropis amabilis occurred at all sites and was abundant (> 2%) at six
sites. Moxostoma conges turn occurred at six sites and was most

190

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

Table 3: Mean seasonal species richness (S), Shannon diversity (H’), and Pielou’s
evenness (J’) for eight Blanco River sites, the Little Blanco River, and Cypress
Creek between October 2003 and July 2005.

Site:

Fall

Winter

J’

Spring

Summer

Site 1

10.5

1.08

0.46

9.5

1.32

0.61

7.0

1.38

0.76

9.0

0.91

0.41

Reservoir

10.0

1.67

0.73

8.5

1.16

0.54

9.0

1.35

0.62

9.5

1.21

0.55

Site 3

10.0

1.48

0.64

9.0

1.30

0.62

9.5

1.21

0.54

12.5

1.38

0.55

Site 4

6.5

1.27

0.70

3.5

0.66

0.53

6.5

1.34

0.72

9.5

1.49

0.68

Site 5

7.0

0.88

0.45

11.0

1.20

0.50

9.0

1.18

0.56

9.0

1.02

0.47

Site 6

7.0

1.03

0.55

8.5

1.25

0.58

11.5

1.37

0.56

11.0

1.25

0.52

Site 7

12.0

0.87

0.35

11.5

1.77

0.73

11.5

1.17

0.48

12.5

1.29

0.51

Site 8

14.0

1.71

0.65

12.0

1.03

0.42

11.0

1.25

0.52

8.0

0.82

0.40

LBR

10.5

1.74

0.74

9.0

1.73

0.79

10.0

1.77

0.77

12.5

2.23

0.88

CypCr

10.0

1.50

0.65

11.0

1.81

0.76

8.5

1.54

0.72

7.5

1.20

0.58

abundant at the Little Blanco River (2.2%) and the Reservoir (0.9%).
Fish initially identified as Micropterus treculii based on morphology
(Hubbs 1991) were present at four sites with a relative abundance <
0.3% at all sites and 0.04% overall. However, subsequent genetic
analyses conducted on the Blanco River population failed to detect
pure M treculii in the population, which suggests that only
Micropterus dolomieu x M treculii hybrids exist in the Blanco River
drainage (Littrell et al. 2007).

Physical parameters, site, and season accounted for 40% {P <0.01)
of the variation in the Blanco River drainage fish assemblage. Pure
effects of physical parameters accounted for 15.3% {P < 0.01), site
accounted for 1 1.2% (P < 0.01), and season accounted for 2.3% (P <
0.01) of fish assemblage variation. Two- and three-way shared effects
among physical parameters, site, and season accounted for 10.7% of
fish assemblage variation. Physical parameters with the strongest
positive centroids for the first canonical axis (CA I) were riffle (1.57),
side channel (0.64), and maximum velocity (0.56). Physical
parameters with the strongest negative centroids for CA I were
reservoir (-1.28), pool (-0.78), and silt (-0.54). Within the mainstem
of the Blanco River, CA I centroids were negative for sites 1 through
3 and positive for sites 4 through 8. Among the tributaries. Little
Blanco River had a negative centroid whereas Cypress Creek had a
positive centroid. CA I expressed a gradient from upstream sites with
slow current velocities, greater depths, silt substrate, and detritus to
downstream sites with faster current velocities, shallower depths, and

BEAN, BONNER & LITTRELL

191

cobble substrate. Physical parameters with the strongest positive
centroids for the second canonical axis (CA II) were riffle (1.00),
backwater (0.54), and detritus (0.27). Physical parameters with the
strongest negative centroids for CA II were sand (-0.72), plunge pool
(-0.70), run (-0.52), and boulder (-0.48). CA II expressed a weaker
habitat gradient from shallow backwaters to deeper runs with sand
substrate. The strongest negative loadings of CA II described the
habitat at Site 5. Summer and fall had negative centroids for CA I and
winter and spring had positive centroids for CA I, however, these
centroids were generally weak.

Species with the strongest positive associations with CA I include
Percina sciera, Pimephales promelas, Percina carbonaria,
Etheostoma spectabile, and Ameiurus natalis (Fig. 4). Species with
the strongest negative associations with CA I include Lepomis
microlophus, Micropterus salmoides, Cyprinus carpio, Cyprinella
lutrensis, and Lepomis gulosus. Along the habitat gradients expressed
by CA I and CA II, P. carbonaria (N = 20) and E. spectabile (N =
540) were strongly associated with riffles having high current
velocities, shallow depths, and intermediate-size substrate such as
gravel and cobble. Cobble and gravel substrates were dominant at
Site 8 where E. spectabile relative abundance was highest among
mainstem sites (Table 2). Cyprinella venusta {N = 11,918) and G.
affinis {N = 2,419) were associated with intermediate currents and
showed no strong substrate affinities. Campostoma anomalum {N =
1,160) and Ictalurus punctatus (A = 110) were associated with
intermediate current velocities, shallow depths, and cobble substrate.
Notropis amabilis {N ^ 3,308) and M dolomieu {N = 33) were
associated with intermediate current velocities and coarse substrates.
Fish species associated with deep, low-velocity habitats with greater
amounts of vegetation and detritus included C. lutrensis {N = 5), C.
carpio {N = 3), D. nigrotaeniata {N ^ 36), P. vigilax (N = 4,136), M
congestum (N = 115), Lepomis cyanellus {N = 165), L. gulosus {N =
6), Lepomis macrochirus (N = 795), Lepomis megalotis (N = 511), Z.
microlophus {N = 10), and M salmoides {N = 196). Species strongly
associated with pool and reservoir habitats included C. carpio, C.
lutrensis, P. vigilax, M. congestum, L. gulosus, L. macrochirus, and L.
megalotis.

192

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

Figure 4. CCA ordination plots of (a) species and (b) physical habitat parameters for the
Blanco River October 2003 through July 2005.

BEAN, BONNER & LITTRELL

193

Although pure site effects explained only 11.2% of fish
assemblage variation, several species showed strong site affinities.
Macrhybopsis marconis (N = 4), Astyanax mexicanus (N = 15),
Poecilia latipinna (N= 1), Lepomis punctatus {N = 16), and P. sciera
(A = 1) occurred exclusively at sites 7 and 8 (Table 2) downstream
from the lowermost low-head dam on the Blanco River and nearest to
the confluence with the San Marcos River. Cyprinus carpio was
collected only from the Reservoir Site and in a deep pool at Site 4.
Etheostoma spectabile, P. carbonaria, and P. sciera were present
downstream of the falls at Site 4 and were absent above the falls.
Dionda nigrotaeniata was present in the tributaries (Little Blanco
River and Cypress Creek) but was absent from mainstem sites.

Sites were grouped into four categories to test for influence of low-
head dams on fish assemblages. Site categories were “adjacent to
low-head dams (<1 km)” (sites 1 and 7); “impounded by a low-head
dam” (Reservoir Site); “distant from a low-head dam (>1 km)” (sites
3, 4, 5, 6, and 8), and “tributaries” (Little Blanco River and Cypress
Creek). Analysis of similarities (Table 4) indicated that differences
existed {P <0.01) among fish assemblages relative to low-head dams.
Pair-wise tests indicated significant differences between impounded
and adjacent site assemblages {P < 0.01), impounded and distant site
assemblages {P = 0.04), adjacent and tributary site assemblages {P <
0.01), and distant and tributary site assemblages {P < 0.01). Fish
assemblage differences were not detected between impounded and
tributary site assemblages (P = 0.07) or between adjacent and distant
site assemblages (P = 0.70). Fish assemblage differences between
impounded and adjacent and between impounded and distant
assemblages were attributed to the large number of P. vigilax (61%)
and lower number of C. venusta (20%) in the Reservoir Site relative
to riverine sites.

Discussion

Physical habitat parameters, site effects, and seasonal effects
explained significant amounts of variation within the Blanco River
fish assemblage. However, the amount of variation explained by
season was small whereas physical habitat parameters explained a

194

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

Table 4: Results of ANOSIM global and pair-wise tests for differences in assemblage
between sites impounded by a low-head dam, adjacent to a low-head dam, distant
from a low-head dam, and tributaries.

P value

Global Test

0.264

<0.01

Pairwise Tests

Near

vs.

Impounded

0.654

<0.01

Near

vs.

Distant

-0.063

0.70

Near

vs.

Tributary

0.585

<0.01

Impounded

vs.

Distant

0.346

0.04

Impounded

vs.

Tributary

0.267

0.07

Distant

vs.

Tributary

0.386

<0.01

relatively large amount of variation. Few species (i.e., N. volucellus,
P. vigilax, G. affinis, and M salmoides) showed strong seasonal
trends in relative abundances. Relative abundance was highest for G.
affinis and N. volucellus in the fall. Mean length was lowest for both
species in fall, thus the higher relative abundances may represent an
abundance of juveniles prior to significant seasonal mortality
(Matthews 1998). The peak in P. vigilax relative abundance occurred
in winter and resulted from a single seine haul at the Reservoir Site.
Micropterus salmoides abundance was highest in summer when age-0
fish were small and unable to escape from seines. This presumably
resulted in a greater capture rate for smaller individuals (Weinstein &
Davis 1980).

The extent of site associations was highly variable among species.
Dionda nigrotaeniata typically inhabits spring-influenced headwaters
(Hubbs et al., 1991) and showed a strong site affinity for the Little
Blanco River and Cypress Creek. Ameiurus natalis was collected
exclusively in Cypress Creek and is often associated with small, clear,
rock- or gravel-bottomed streams (Robison & Buchanan 1988). All
percids (i.e., E. spectabile, P. carbonaria, and P. sciera) were absent
above Site 4. This observation is consistent with reports by the Texas
Game and Fish Commission (now Texas Parks and Wildlife; 1957),
indicating that the falls at Site 4 present an apparent barrier to the
upstream movement of fishes, and contribute to an abrupt
discontinuity in assemblage variation. Macrhybopsis marconis, A.

BEAN, BONNER & LITTRELL

195

mexicanus, P. latipinna, and P. sciera occurred only at sites 7 and 8.
These fishes were rare, and likely represent vagrants from the San
Marcos River, as larger, more stable water bodies serve as species
pools from which less stable upstream habitats are colonized
(Whiteside & McNatt 1972). Spatial variation within the Blanco
River fish assemblage across geographically distant sites (i.e.,
tributary vs. mainstem and upstream vs. downstream) likely represents
differences in stream processes (Wilkinson & Edds 2001). The
relatively small amounts of spatial and seasonal variation suggest
adequately stable habitats with assemblages primarily structured by
local habitat parameters (Meador & Matthews 1992).

Physical habitat parameters were found to be the primary factors
structuring fish assemblages. The first two canonical axes of CCA
both represented gradients best described as velocity and substrate
gradients. Along these gradients, centrarchids were generally more
abundant in lentic type habitats with lower velocities and greater
percentages of silt substrate and vegetation whereas percids were
more abundant in shallow lentic habitats such as riffles dominated by
cobble and gravel substrate. Cyprinids, however, exhibited a much
wider range of habitat associations. For example, C. anomalum was
strongly associated with shallow riffles whereas N. amabilis was
associated with runs and C. venusta did not exhibit any strong
associations. Gorman & Karr (1978) noted that stream depth, current
velocity, and substrate are important in structuring stream fish
assemblages. Greater variability in these components results in
increased habitat complexity which regulates local assemblages of
fish as stream fish are commonly habitat specialists (Mendelson
1975).

The low-head dam at the Reservoir Site created a distinctly lentic
habitat in which centrarchids were abundant, as were the ubiquitous
C. venusta and P. vigilax. Santucci et al. (2005) reported major
differences in habitat quality between impounded and free flowing
reaches such as higher turbidity, lower dissolved oxygen minima (as
low as 2.5 mg/L), and homogenization of habitats. Homogenization
of habitats can lead to increased abundance of generalist native
species (Scott & Helfman 2001) and loss of native stream specialists
(Boet et al. 1999). In addition to habitat alterations, fish assemblages

196

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

within reservoirs may be altered by purposeful introductions of
sportfish species as well as incidental introductions via bait-buckets
(Taylor et al. 2001). The Reservoir Site ranked sixth in species
richness with 17 species collected, while the mean species richness
across all sites was 17.3 species. Although species richness at the
Reservoir Site was nearly average for the sites, the structure of the
assemblage at this site was significantly different from that of sites
both adjacent to and distant from reservoirs. Campostoma anomalum,
M. dolomieu, and M treculii, which typically inhabit swifter waters,
were rare or absent from the Reservoir Site, but were present and
more abundant at Site 3. Gillette et al. (2005) and Taylor et al. (2001)
reported similar shifts in assemblages with lower abundances of fishes
normally associated with higher current velocities, such as percids and
stream dwelling micropterids. Additionally, the absence of percids at
the Reservoir Site is attributed to the natural barrier present at Site 4
and is likely not the result of the low-head dam.

The Little Blanco River consisted mostly of a series of pools due to
low-flow conditions during the study, and was dominated by lentic
species (i.e., centrarchids). In contrast, the consistently flowing
Blanco River and Cypress Creek were dominated by cyprinids.
Periods of low flow and increased centrarchid abundance likely
resulted in the relatively high degree of similarity between tributaries
and the Reservoir and a lack of a significant difference between the
two treatment groups in analysis of similarities. Although no
differences were detected among assemblages at sites either adjacent
to or distant from impoundments, it cannot be concluded that low-
head dams do not impact the fish assemblage as there is no
comparison to the pre- impoundment assemblage structure.

Although no pre-impoundment assemblage data are available,
several changes in occurrence have taken place since the Texas Parks
and Wildlife Department survey in 1957. Dorosoma cepedianum,
Carpiodes carpio, Notemigonus crysoleucas, Notropis buchanani,
Ameiurus melas, Pomoxis annularis, and Etheostoma lepidum were
not collected in our sampling efforts, but were present, though
relatively rare, in the 1957 collection. Species present in our
collections but absent from the 1957 collection include the introduced

BEAN, BONNER & LITTRELL

197

species P. promelas, P. latipinna, and M dolomieu, as well as P.
sciera which is likely a vagrant from the San Marcos River.

Physical habitat parameters, site, and season all influenced Blanco
River fish assemblage. However, there is a substantial amount of
variability in the assemblage that is not explained by the present
model. This variability may reflect unmeasured or inestimable factors
such as land use, riparian vegetation, or biological interactions that act
to structure the fish assemblage at the local and watershed level.
Although these unmeasured factors are likely to influence the
structure of the fish assemblage, current velocity, depth, and substrate
often adequately account for variation in fish assemblages at a local
scale (Gorman & Karr 1978; Schlosser 1982; Cantu & Winemiller
1997; Walters et al. 2003; Williams et al. 2005). Such factors,
comprising major aspects of stream morphology, are likely to shift in
response to anthropogenic disturbance (Odemerho 1984; Golladay et
al. 1987), and the fish assemblage can be expected to track with such
shifts.

Among the threats to the Blanco River and other Texas Hill
Country streams, excessive groundwater pumping is the greatest
(Bowles & Arsuffi 1993). Continued excessive pumping, especially
during drought, will likely result in a loss of spring associated
headwater specialists such as D. nigrotaeniata. As rapid urbanization
continues, changes in geomorphology and hydrology will likely
include increased impervious cover resulting in higher but shorter-
duration hydrograph peaks and changes in substrate composition
resulting from siltation, similar to changes in nearby and urbanized
Waller Creek in Austin, Texas (Swezey 1991). Such changes in
stream characteristics will likely influence changes in fish assemblage
structure as several species in the Blanco River are strongly associated
with particular substrate types (e.g., percids).

Acknowledgments

The River Systems Institute at Texas State University, The Nature
Conservancy of Texas, and Peter Way (Way Ranch) provided partial
funding for this project. We thank Megan Bean, Casey Williams, and

198

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

Jackie Watson for assistance in field collections and Peter Way, Hall
and Pat Hammond, Jerry and Diane Turner, Dorothy Gumbert, Bill
Buchanan, and Don and Nelle Still for river access. A. W. Groeger
and D. G. Huffman provided comments on earlier versions of this
manuscript.

Literature Cited

Adams, W. M. 1985. The downstream impacts of dam construction: a case study from
Nigeria. Transactions of the Institute of British Geographers, 10:292-302.

Arend, K. K. 1999. Classification of streams and reaches. Pages 57-74, in M. B. Bain
and N. J. Stevenson, editors. Aquatic habitat assessment: common methods.
American Fisheries Society, Bethesda, Maryland, 224 pp.

Boet, P., J. Belliard, R. Berrebi-dit-Thomas & E. Tales. 1999. Multiple human impacts
by the city of Paris on fish communities in the Seine River basin, France.
Hydrobiologia, 410:59-68.

Bonner, T. H. & G. R. Wilde. 2000. Changes in the Canadian River fish assemblage
associated with reservoir construction. Journal of Freshwater Ecology, 15:189-198.

Borcard, D., P. Legendre & P. Drapeau. 1992. Partialling out the spatial component of
ecological variation. Ecology, 73:1045-1055.

Bowles, D. E. & T. L. Arsuffi. 1993. Karst aquatic ecosystems of the Edwards Plateau
region of central Texas, USA: a consideration of their importance, threats to their
existence, and efforts for their conservation. Aquatic Conservation: Marine and
Freshwater Ecosystems, 3:317-329.

Brown, J. H. & B. A. Maurer. 1989. Macroecology: the division of food and space
among species on continents. Science, 243:1 145-1 150.

Burr, B. M. & R. L. Mayden. 1992. Phylogenetics and North American freshwater
fishes. Pages 18-75, in: R. L. Mayden, ed. Systematics, historical ecology, and
North American freshwater fishes. Stanford University Press, Stanford, CA., 969 pp

Cantu, N. E. V. & K. O. Winemiller. 1997. Structure and habitat associations of Devils
River fish assemblages. Southwestern Naturalist, 42:265-278.

Cave, M. 2006. Effects of surface and groundwater interactions on the solution
chemistry of a subtropical karst stream. Unpublished Masters Thesis. Texas State
University-San Marcos, 43 pp.

Clarke, K. R. & R. H. Green. 1988. Statistical design and analysis for a ‘biological
effects’ study. Marine Ecology Progress Series, 46:213-226.

Clarke, K. R. 1993. Non-parametric multivariate analyses of changes in community
structure. Australian Journal of Ecology, 18:117-143.

Edwards, R. J. 1978. The effect of hypolimnion reservoir releases on fish distribution
and species diversity. Transactions of the American Fisheries Society, 107:71-77.

Gillette, D. P., J. S. Tiemann, D. R. Edds & M. L. Wildhaber. 2005. Spatiotemporal
patterns of fish assemblage structure in a river impounded by low-head dams.
Copeia, 2005:539-549.

Golladay, S. W., J. R. Webster & E. F. Benfield. 1987. Changes in stream morphology
and storm transport of seston following watershed disturbance. Journal of the North
American Benthological Society, 6:1-1 1.

BEAN, BONNER & LITTRELL

199

Gorman, O. T. & J. R. Karr. 1978. Habitat structure and stream fish communities.
Ecology, 59:507-515.

Harding, J. S., E. F. Benfield, P. V. Bolstad, G. S. Helfman & E. B. D. Jones. 1998.
Stream biodiversity: the ghost of land use past. Proceedings of the National
Academy of Sciences of the United States of America, 95:14843-14847.

Herbert, M. E. & F. P. Gelwick. 2003. Spatial variation of headwater fish assemblages
explained by hydrologic variability and upstream effects of impoundment. Copeia,
2003:273-284.

Hubbs, C., R. J. Edwards & G. P. Garrett. 1991. An annotated checklist of the
freshwater fishes of Texas, with keys to identification of species. Texas Journal of
Science Supplement, 43:1-56.

Krebs, C. J. 1999. Ecological methodology, 2"^* ed. Addison-Welsey Educational
Publishers, Inc., Menlo Park, California, 620 pp.

Littrell, B. M., D. Lutz-Carrillo, T. H. Bonner & L. T. Fries. 2007. Status of an
introgressed Guadalupe bass population in a central Texas stream. North American
Journal of Fisheries Management, 27:785-791.

Matthews, W. J. 1998. Patterns in freshwater fish ecology. Chapman & Hall, New
York, N.Y„ 784 pp.

Meador, M. R. & W. J. Matthews. 1992. Spatial and temporal patterns in fish
assemblage structure of an intermittent Texas stream. American Midland Naturalist,
127:106-114.

Mendelson, J. 1975. Feeding relationships among species of Notropis (Pisces:
Cyprinidae) in a Wisconsin stream. Ecological Monographs, 45:199-232.

Odemerho, F. O. 1984. The effects of shifting cultivation on stream channel size and
hydraulic geometry in small headwater basins of southwestern Nigeria. Geografiska
Annaler. Series A, Physical Geography, 66:327-340.

Porto, L. M., R. L. McLaughlin & D. L. G. Noakes. 1999. Low-head barrier dams
restrict the movements of fishes in two Lake Ontario streams. North American
Journal of Fisheries Management, 19:1028-1036.

Richter, B. D., D. P. Braun, M. A. Mendelson & L. L. Master. 1997. Threats to
imperiled freshwater fauna. Conservation Biology, 1 1 : 1 08 1 - 1 093 .

Robison, H. W. & T. M, Buchanan. 1988. Fishes of Arkansas. University of Arkansas
Press, Fayetteville, AR, 544 pp.

Santucci, V. J., Jr., S. R. Gephard & S. M. Pescitelli. 2005. Effects of multiple low-head
dams on fish, macroinvertebrates, habitat, and water quality in the Fox River, Illinois.
North American Journal of Fisheries Management, 25:975-992.

Schlosser, 1. J. 1982. Fish community structure and function along two habitat gradients
in a headwater stream. Ecological Monographs, 52:395-414.

Scott, M. C. & G. S. Helfman. 2001. Native invasions, homogenization, and the
mismeasure of integrity of fish assemblages. Fisheries, 26:6-15.

Shmida, A. & M. V. Wilson. 1985. Biological determinants of species diversity.
Journal of Biogeography, 12:1-20.

Shuman, J. R. 1995. Environmental considerations for assessing dam removal
alternatives for river restoration. Regulated Rivers: Research & Management,
11:249-261.

Swezey, C. S. 1991. A review of changes in the geomorphology and hydrology of
Waller Creek (Austin, Texas) as a result of urban development. Texas Journal of
Science, 43:315-323.

200

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

Taylor, C. A., J. H. Knouft & T. M. Hiland. 2001. Consequences of stream
impoundment on fish communities in a small North American drainage. Regulated
Rivers: Research & Management, 17:687-698.

ter Braak, C. J. F. 1986. Canonical correspondence analysis: a new eigenvector
technique for multivariate direct gradient analysis in ecology. Ecology, 67:1167-
1179.

Texas Game and Fish Commission. 1957. Basic survey and inventory of fish species
present, as well as their distribution in the Blanco River, its tributaries and watershed
lying within Blanco, Kendall and Hays Counties, Texas. Project No. F9R3, Job B-

10, 26 pp.

United States Geological Survey. 2003. Water Resources Data-Texas, Water Year 2002,
Volume 5 Guadalupe River, Nueces River, Rio Grande River Basins and Intervening
Coastal Basins. Water Data Report TX-02-5, 414 pp.

Walters, D. M., D. S. Leigh, M. C. Freeman, B. J. Freeman & C. M. Pringle. 2003.
Geomorphology and fish assemblages in a Piedmont river basin, U.S.A. Freshwater
Biology, 48:1950-1970.

Warren, M. L., Jr., B. M. Burr, S. J. Walsh, H. L. Bart, Jr., R. C. Cashner, D. A. Etnier,
B. J. Freeman, B. R. Kuhajda, R. L. Mayden, H. W. Robison, S. T. Ross & W. C.
Starnes. 2000. Diversity, distribution, and conservation status of the native
freshwater fishes of the southern United States. Fisheries, 25:7-31.

Warren, M. L., P. L. Angermeier, B. M. Burr & W. R. Haag. 1997. Decline of a diverse
fish fauna: patterns of imperilment and protection in the southeastern United States.
Pages 105-164, in G. W. Benz and D. E. Collins, editors. Aquatic fauna in peril: the
southeastern perspective. Special Publications 1, Southeast Aquatic Research
Institute, Lenz Design and Communications, Decatur, GA, 554 pp.

Weinstein, M. P. & R. W. Davis. 1980. Collection efficiency of seine and rotenone
samples from tidal creeks. Cape Fear River, North Carolina. Estuaries, 3:98-105.

Whiteside, B. G. & R. M. McNatt. 1972. Fish species diversity in relation to stream
order and physicochemical conditions in the Plum Creek drainage basin. American
Midland Naturalist, 88:90-101.

Wilkinson, C. D. & D. R. Edds. 2001. Spatial pattern and environmental correlates of a
Midwestern stream fish community: including a spatial autocorrelation as a factor in
community analysis. American Midland Naturalist, 146:271-289.

Williams, C. S. & T. H. Bonner. 2006. Habitat associations, life history, and diet of the
Sabine shiner Notropis sabinae in an East Texas drainage. American Midland
Naturalist, 155:84-102.

Williams, J. E., J. E. Johnson, D. A. Hendrickson, S. Contreras-Balderas, J. D. Williams,
M. Navarro-Mendoza, D. E. McAllister & J. E. Deacon. 1989. Fishes of North
America endangered, threatened, or of special concern: 1989. Fisheries, 14:2-20.

Williams, L. R., T. H. Bonner, J. D. Hudson III, M. G. Williams, T. R. Leavy & C. S.
Williams. 2005. Interactive effects of environmental variability and military
training on stream biota of three headwater drainages in western Louisiana.
Transactions of the American Fisheries Society, 134:192-206.

Winston, M. R., C. M. Taylor & J. Pigg. 1991. Upstream extirpation of four minnow
species due to damming of a prairie stream. Transactions of the American Fisheries
Society, 120:98-105.

PTB at: preston.bean@txstate.edu

TEXAS J OF SCI. 59(3):20 1-208

AUGUST, 2007

DISTRIBUTION OF CAGLE^S MAP TURTLE (GRAPTEMYS CAGLEI)
IN THE BLANCO AND SAN MARCOS RIVERS

Thomas R. Simpson and Francis L. Rose

Wildlife Ecology Program, Department of Biology,

Texas State University-San Marcos,

San Marcos, Texas 78666

Abstract -Cagle’s Map Turtle {Graptemys caglei) is a small emydine aquatic
turtle described in 1974. At that time, it was known from the Guadalupe and San
Antonio river systems, including the Guadalupe, San Antonio, Blanco, and San
Marcos rivers. Its presence in the Blanco and San Marcos rivers was based on few
documented specimens. Surveys in 1991 and 1992 failed to detect the turtle’s
presence in the Blanco River or the upper reaches of the San Marcos River.
Grapemys caglei was listed as threatened by Texas Parks and Wildlife in 2000 and is
classified as vulnerable by the lUCN Red List due to the diminished distribution. In
2003, a male and female G. caglei were found at one location in the Blanco River.
Systematic searches begun in 2005 revealed small-scattered populations of G. caglei
in the Blanco River from near Fisher to San Marcos in Hays County. No G. caglei
were observed in the. upper reaches of the San Marcos River (Hays County);
however, they were observed at several sites on the San Marcos River in Guadalupe
and Gonzalez counties. Extensive surveys and trapping efforts over 15 years resulted
in 4,000 individually marked turtles in Spring Lake (Hays County), the initial source
for the San Marcos River, and no Graptemys were observed. The river dynamic
(extreme drought to extreme flood) of the Blanco River probably plays a dominant
role in the turtle’s life history. This dynamic is muted in the lower San Marcos and
the Guadalupe rivers

Raun (1959) reported two juvenile and one subadult Graptemys
pseudogeographia from the San Marcos River near Ottine,
Gonzales County, Texas. Members of the genus Graptemys, at that
time, were not known to occur west of the Colorado River. Proper
species allocation of these three turtles was possible when Cagle’s
map turtle {Graptemys caglei) was described by Haynes &
McKown (1974) from the Guadalupe River system in south-central
Texas. The new species was distinguished from other members of
the genus by its transverse yellowish chin bar and a conspicuous
“V” dorsal head marking. It belongs to the narrow-headed group of
map turtles (Dobie 1981). Although extensively studied, little has
been published about its natural history. It is an inhabitant of the
Guadalupe River where it may be the most common species of

202

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

aquatic turtle in DeWitt County (Vermersch 1992). It is known to
occur upriver to the Kerrville area in Kerr County, but its
distribution in the Guadalupe River across the Edwards Plateau is
poorly documented. Haynes & McKown (1974) report a specimen
from 15 km NW of San Marcos, which is in the Blanco River (Hays
County); however. Porter (1992) found no Cagle’s map turtles on
the Blanco River during his surveys and stated, “Cagle’s map
turtles seem to be disappearing from areas previously reported to
support populations.” A review of the literature confirms that actual
sightings of this turtle in the Blanco River were minimal and
Killibrew et al. (2002) stated that the turtle “...is currently found
only in segments of the Guadalupe and San Marcos rivers...”
Because of its presumed diminished distribution, the Texas Parks
and Wildlife Division (2000) classified the turtle as “threatened”
and it was listed as “vulnerable” by the lUCN Red List (Tortoise &
Freshwater Turtle Specialist Group 1996).

A male and female Graptemys caglei were captured by hand
within 10 minutes of each other at one location in the Blanco River,
Hays County, in 2003. Because of these two individuals a
systematic search for the turtle in the Blanco and San Marcos rivers
was begun in 2005. More intense searches of the San Marcos River
were begun in 2006. The results of those systematic searches are
presented in this report.

Methods and Materials

Initial searches for Graptemys caglei suggested that the
populations were dispersed. Therefore, systematic searches of
selected sites for presence or absence of G. caglei was the chosen
method. Search sites were selected based on access and the
presence of optimal habitat characteristics for G. caglei, including
riffles, shallow pools, moderate water flow and sand or gravel bars
(Haynes & McKown 1974; Killebrew 1991; 1992; Babitzke 1992).
Searches were conducted on foot or from kayaks. Dip nets were
used to capture basking or swimming turtles. Search site
coordinates were recorded with a GPSMAP 60cs unit (Garmin
International, Inc., Olathe, KS).

SIMPSON & ROSE

203

For identification, dorsal head and cheek patterns, wide forearm
and post femoral yellow stripes, and ventral chin patterns were
considered definitive for Cagle's map turtles. The primary head and
cheek patterns are: (a) post orbital stripe forming a crescent
beginning underneath the eye and extending vertically upward on
the side of the head (Figure la), (b) v-shaped pattern formed by
medio-dorsal extensions of the post-orbital stripe on top of the
head, and (c) unbroken transverse crescent underneath the chin
(Figure lb.) (Haynes & McKown 1974). The Texas river cooler
(Pseudemys texana) is the only other species with which G. caglei
might be confused within an appropriate size range. Typical P.
texana head markings in the area of sympatry include a large
yellow post-orbital spot, bold supralabial and infralabial stripes, and
numerous ventral medial stripes. However, head patterns of P.
texana are geographically variable. Whereas, juveniles of the two
species are morphologically similar, the three juvenile G. caglei
that were observed during the study were distinctive, with brightly
colored, wide yellowish carapacial markings. Vertebral “knobs”
are not definitive because juvenile P. texana can have prominent
extensions that may be tipped in black and the more prominent
“knobs” of G caglei are reduced ontogenetically. Observers
adhered to the general rule: If there is confusion about the visual
identification, the species at hand is P. texana.

Results

Blanco i?/ver. -Initially, two Graptemys caglei were found on a
private ranch in the Blanco River (29° 59'N, 97° 58'W) nine km
west of Kyle by students in a field class in 2003. During 2005 and
2006, 453 search hours were conducted at 10 sites. Graptemys
caglei was identified from four sites on the Blanco River. One
individual was seen in a pool (29° 54'N, 97° 53'W), 8.2 km above
the confluence of the Blanco and San Marcos rivers. Two adults
and 3 juveniles were visually identified at Dudley Johnson Park,
Hays County (29° 57'N, 97° 54'W), 13.7 km above the confluence
with the San Marcos River. One small individual (83 mm carapace

204

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

SIMPSON & ROSE

205

length) was captured -30 km upstream of the confluence of the
Blanco and San Marcos Rivers (30° OO'N, 97° 58'W). One large
female (158 mm carapace length) was captured and five adults were
observed 3.5 km downstream (29° 58'N, 98° 12'W) of the Fischer
School Road crossing, Blanco River, Hays County. No individuals
of G. caglei were found in Blanco State Park (30° 05N, 98° 25'W)
in Blanco, nor within 4 km upstream or downstream of the park.
Visits to sites upstream of Blanco State Park revealed little habitat
compatible with turtles.

San Marcos -Numerous search and trapping efforts have
been done in Spring Lake (29° 53'N, 97° 56'W) and the upper San
Marcos River within San Marcos, Hays County, from 1993 to
present. Although, more than 4000 turtles were captured and
marked at these sites, no Graptemys caglei individuals were seen or
captured.

Searches for Graptemys caglei on the San Marcos River were
oriented to road and highway crossings in most instances. Each of
17 road and highway crossings between San Marcos and Gonzalez
were visited and searched a minimum of three times for a total of
110 search hours. Map turtles were observed at four of these sites.
Four G. caglei individuals were identified visually with spotting
scopes at Farm to Market (F.M.) 20, Gonzalez County (29° 45N,
97° 46W). Two G. caglei individuals were identified at F.M. 2091
(29° 35’N, 97° 35'W), Ottine, Texas. One individual of G. caglei
was visually identified at the crossing of County Road 232 (29°
33’N, 97° 32'W) and at U. S. Highway 90a (29° 30’N, 97° 29’W).

Discussion

The Blanco River presents several challenges to aquatic
vertebrates: (1) floods frequently alternate with severe droughts
during which there is no flow, (2) the substrate is limestone based
and during the summer, higher temperatures in shallow water limit
the organisms that can live there, and, (3) the scoured substrate has
limited attached aquatic plants and few invertebrates, including

206

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

snails and young clams, that could serve as food. The most
numerous chelonians observed at all search sites in the Blanco
River were the red-eared slider (Trachemys scripta) and Pseudemys
texana, in roughly equal numbers. However, densities of all turtle
species were low compared to numbers observed on the San
Marcos River. Results of this study support the summation that
whereas Graptemys caglei does inhabit the Blanco River, its
densities are low, and the turtles are probably restricted to areas
where deep pools provide suitable habitat during drought. Basking
and nesting sites might also be limiting.

No Graptemys caglei were observed in or upstream of Lake
Blanco (Blanco County). However, significantly more turtles were
observed within the lake, including the common musk turtle
(Sternotherus odoratus) and the spiny softshell turtle (Apalone
spinifera). The negative environmental factors that probably limit
turtle distributions in the Blanco River are accentuated upstream of
Blanco Lake, especially during times of drought. In addition, the
area between Dudley Johnson Park (Hays County) and the
confluence with the San Marcos River is also deemed
environmentally challenging for G. caglei for the same reasons.
Distributions downstream might be enhanced by floods; however,
once water levels recede, habitat quality for maintenance of healthy
G. caglei populations is limited.

The San Marcos River has significant flow and has not gone dry
in recorded history (Brune 2000). Turtle densities are high, yet the
presence of G. caglei has been documented infrequently in its upper
reaches, and there are no records in the San Marcos River above its
confluence with the Blanco. There are several substantial dams on
the San Marcos River above Interstate Highway 10. The three
largest, Luling City Park Dam (Luling), Martindale Dam
(Martindale), and Cummings Dam near the confluence of the San
Marcos and Blanco rivers), probably limit upstream migration. In
addition, the deep water produced by these dams is not thought to
be optimal habitat for G. caglei populations (Killibrew, et al. 2002).

SIMPSON & ROSE

207

Although these surveys confirm that there are viable local
populations of G. caglei in the Blanco River, it should be
emphasized that these are scattered and densities are low (about 10
turtles/453 search hours). This confirmation does not diminish the
need for continued surveillance and protection.

Acknowledgments

We thank S. Franklin, J. Scalise, and R. Swanson for their help
in the field and G. Amoan, P. Way, and W. Johnson for access to
segments of the Blanco River. We also thank the Texas Nature
Conservancy, Texas Rivers Center, and Texas Department of
Transportation for funding. Permits were provided by Texas Parks
and Wildlife (permit number SPR-0993-638) and Texas State
University Institutional Animal Use and Care Committee (permit
number 06-06325 15E7F).

Literature Cited

Babitzke, J. B. 1992. An analysis of population size of Graptemys caglei.
Unpublished Masters Thesis, 47 pp.

Brune, G, 2002. Springs of Texas, Vol 1. 2002. Texas A&M University Press,
College Station, Texas, 566 pp.

Dobie, J. L. 1981. The taxonomic relationships between Malaclemmys Gray, 1 844
and Graptemys Agassiz, 1857 (Testudines: Emydidae). Tulane Studies in
Zoology and Botany, 23:85-102.

Haynes, D. & R. R. McKown. 1974. A new species of map turtle (Genus
Graptemys) from the Guadalupe River System in Texas. Tulane Studies in
Zoology and Botany, 18:143-152.

Killibrew, F. C. 1991, A petition for threatened status listing of Graptemys caglei
(Testudines, Emydidae). Letter to U.S. Fish and Wildlife Service, Corpus Christi
Ecological Services Field Office. April 8, 1991, 15 pp.

Killibrew, F. C, 1992. U.S. Fish and Wildlife Service scope. A synopsis of
information on Graptemys caglei. Unpublished report for U.S. Fish and Wildlife
Service, Austin, Texas, 10 pp.

Killibrew, F. C., W. J. Rogers & J. B. Babitzke. 2002. Assessment of instream flow
and habitat requirements for Cagle’s map turtle {Graptemys caglei). Report to
Edwards Aquifer Authority, contract #00-52-AS, 60 pp.

Porter, D. A. 1992. Distribution survey on Graptemys caglei. Unpublished report to
U. S. and Wildlife Service, Austin, TX, 6 pp.

Raun, G. G. 1959. Terrestrial and aquatic vertebrates of a moist, relict area in
Central Texas. Texas Journal Science, 1 1(2):158-17L

208

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

Tortoise & Freshwater Turtle Specialist Group. 1996. Graptemys caglei, In: lUCN
2006. 2006 lUCN Red List of Threatened Species, <www.iucnredlist.org>.

Downloaded on 15 January 2007.

Vermersch, T. G. 1992. Lizards and Turtles of South-Central Texas. Eakin Press,
Austin, 170 pp.

TRS at: r_simpson@txstate.edu

TEXAS J. OF SCI. 59(3);209-232

AUGUST, 2007

MODELING FUTURE FLOWS IN THE BLANCO RIVER WATERSHED
UNDER VARIOUS DEVELOPMENT AND RAINFALL SCENARIOS

Joanna C. Curran

Department of Geography, Texas State University
San Marcos, Texas 78666
Current address:

Department of Civil and Environmental Engineering, University of Virginia
Charlottesville, Virginia 22904

Abstract.-The Blanco River of central Texas is supplied through spring flow and
tributaries. Land use in the watershed is predominantly rural, but in recent years the rate
of urbanization within the watershed has increased. Rates of population growth and
urbanization are expected to increase into the future. The SWAT (Soil and Water
Assessment Tool) model is used to predict the impact of land management practices on
water yields in complex watersheds under varying soils, land use, and management
conditions. Using SWAT, the flows in 38 separate sub-basins of the Blanco watershed
are predicted under a range of future land use and climate scenarios. Two climate
eonditions representing current precipitation patterns and drought conditions were
modeled. Three land use urbanization scenarios were modeled; current land use
conditions of 2% urban area, 13% urbanization of the watershed, and 78% urbanization
of the watershed area. Results show a decrease in river flow as the urban area reaches
13% of total land use. This reflects an increase in household water use yet enough rural
area to allow for rainfall infiltration to the subsurfaee. At 78% urbanization of the
watershed, flows to the river are increased. Rainfall does not infiltrate due to the
extensive impervious cover but rather flows directly to the Blanco River as overland run¬
off. Baseflow is maintained through increased discharges from munieipal wastewater
plants. The predicted flows under all seenarios modeled are lowest for the first portion of
the watershed, upstream of any major tributary contributions.

The connection between land use in a watershed and the water
quality and quantity of the main channel has become part of the public
conscience as populations in the arid and semi-arid regions of the US
increase. With the passage and implementation of the Clean Water
Act, water quality issues associated with land use were brought into
the public eye. Over the past 30 years, this awareness has been
extended as the public has become increasingly aware of the possible
threats to natural ecosystems posed by land use changes associated
with urbanization.

As cities plan for expected growth, they often try to choose a
growth strategy that will minimize negative impacts to the
environment. A number of methods are exemplified through a

210

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

sampling of city growth strategies in Texas (Vaughn, pers. comm.).
In Austin in the 1970’s, the city decided to limit hiture growth to a
north-south corridor along 1-35, in part to preserve the recharge area
for the Edwards Aquifer as undeveloped land. The plan was
unsuccessful, as private developers created Municipal Urban Districts
outside the city’s planned growth areas. Over the same period of
time, San Antonio allowed growth in any direction but tried to
influence the growth through zoning laws.

Public involvement in growth planning has increased since the
1970’s, both formally and informally. Formal public participation
exists through public city council meetings and is not new. Informal
participation through events like the Envision Central Texas Project is
a new way in which the public hopes to influence policy. Envision
Central Texas has been holding public meetings in towns throughout
the region between (and inclusive of) Austin and San Antonio where
the attendees are asked to draw their desired type of and areas of
development on a regional map (White Lion & EnviroMedia 2006).
The overwhelming majority of participants placed a high value on
environmental integrity and the preservation of rural space on those
lands where there is recharge to aquifer and stream systems. Through
this exercise, the public demonstrated an awareness and appreciation
for the impact of changing land use within a large watershed area on
the quality and quantity of water in the main channel of the watershed.

Concerns regarding predicted future growth have prompted interest
in the Blanco Watershed of Central Texas. Historic land use in the
watershed is predominantly ranching although there are two small
urban centers: Kyle and Wimberley. Over the past decade, the
growth of these two cities has exploded along with the larger central
Texas region. With more developments planned into the foreseeable
future, the rate of growth of both Kyle and Wimberley is expected to
continue and possibly increase (Rodriguez 2004). As more rural land
is converted to urban use, concerns have arisen about the potential
impact on the Blanco River.

Computer models that simulate the effects of management
decisions over a long time frame are increasingly used to meet the

CURRAN

211

public demand for environmentally responsible growth planning. The
Soil and Water Assessment Tool (SWAT) model is a river basin scale
model that predicts the impact of land management practices on water
yields in complex watersheds with varying soils, land use, and
management conditions. It is a physically based model designed to
predict the impacts to basins without gages or with limited available
data.

This project applies the SWAT model to six possible management
and climate scenarios in the Blanco watershed. One of the predictions
of climate change for the central Texas region is an increased
likelihood of droughts. Thus, two climate periods are modeled. The
first represents current weather conditions and the second simulates
drought conditions. Three management scenarios are modeled:
current conditions, partial watershed urbanization, and majority
watershed urbanization. This paper presents the results of the
modeling efforts.

Study Site

The Blanco River headwaters are springs in northwestern Kendall
County, near the Gillespie County line. The main channel of the
Blanco River flows southeast for 140 km (87 mi), where it joins with
the San Marcos River (McCord 2006). Although the Blanco is a sub¬
watershed of the larger Guadalupe River, its watershed area is 1365
km^ (527 mi^) (Jasinski 2006). As the river flows downstream,
tributaries include Callahan Branch, Flat Creek, the Little Blanco
River, Cypress Creek, and Sink Creek. This study examines the
length of the Blanco River and its tributaries.

The Texas Hill Country features karst topography where limestone
bedrock erodes to form slab benches and a stair-stepped topography.
Channel substrate varies between bedrock and thin layers of alluvium.
Vegetation in undeveloped areas of the watershed is eommonly a mix
of oak, juniper, mesquite, and grasses. Riparian areas are dominated
by bald cypress, oak, elm, cedar, and conifers. The climate supporting
the vegetation is subtropical with short, cool winters and long, hot
summers. Annual precipitation averages near 75 cm but falls
primarily as brief, localized thunderstorms (Jasinski 2006).

212

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

The combination of infrequent rainfall and high temperatures
results in a situation where annual evapotranspiration rates exceed
annual precipitation rates (Short et al. 1987). There is little to no
canopy coverage over the river to reduce evaporation losses. Streams
in the hill country, including the Blanco River, are often intermittent
for all or a portion of their length. In the headwaters, near the springs,
channel flow is perennial. Further downstream, where the Blanco
becomes more dependent on surface runoff, the channel is frequently
intermittent during summer and fall.

Throughout its course, the Blanco maintains a fairly narrow
channel with shallow flow. Channel width varies between 3 and 25
meters (9.8 and 82 feet). At normal flows, the water depth is less than
30 cm (12 inches). Flow in the river is affected by numerous small
run-of-the-river dams placed by local landowners. These dams do not
control flow rates in the channel, but do create small ponds. Two
uses gauges are located within the watershed: at Kyle and
Wimberley. These gauges provide continuous flow records dating
back to 1956 at the Kyle gauge and 1924 at the Wimberley gauge.
The Kyle gauge is located at the outlet of the Blanco watershed and
the Wimberley gauge is situated approximately half-way between the
headwaters and outlet. Data from these gauges are used both as input
to the model and for calibration purposes.

Swat Model

SWAT was designed to predict the impact of topography, soils,
land use, management decisions, and weather on water, sediment,
nutrient, and agricultural chemical yields for large ungaged
watersheds (Srinivasan et al. 1998). A unique feature of SWAT is its
ability to perform continuous time simulations. Many watershed scale
models are capable of simulating only single events, like an individual
storm flow. SWAT has the ability to simulate hundreds of years in a
continuous time step mode, thereby predicting the long-term impacts
of management decisions.

By using a daily time step for simulations rather than the more
common hourly time step, SWAT is able to provide a continuous

CURRAN

213

temporal simulation of physical watershed processes. In a continuous
simulation model, the user can synthesize a database that is analogous
to one created through the collection of continuous data at every sub¬
watershed in the Blanco catchment basin. It would be as if a person
or instrument were present at each of the sub-basin outlets. The
continuous simulation approach automatically takes into account the
serial correlation present in flows and other variables, as well as cross¬
correlations between measured variables when real data are used.
This is potentially the most powerful method available for accurate
prediction of the frequency of receiving water volumes. The
disadvantages associated with this type of model are the large volume
of data required in running it and the inability of the model to predict
specific storm events. The SWAT model is designed to predict data
trends and not specific flows on specific days. It is not a single event
model but rather a model designed to show the implications of large
scale watershed management decisions. The focus of this study is
long-term changes to the watershed due to urbanization with the larger
goal of informing land use management decisions. Thus, the SWAT
model was chosen for this research.

SWAT was designed to be a basin scale model that would include
spatial detail, be computationally efficient, require only readily
available data input, operate over continuous time, simulate user-
specified management scenarios, model ungaged basins, and provide
reasonable results (Arnold & Fohrer 2005). To accomplish this,
SWAT makes use of internal databases that are a collection of
information for approximately 2150 hydrologic areas or Hydrologic
Catalog Units which include data on historical weather, soil
properties, topography, natural vegetation, cropped areas, irrigation,
state and county boundaries, reservoir characteristics, and agricultural
practices. Spatial datasets include topography, land use, soils,
watershed boundaries, stream networks, weather station locations,
aquifer boundaries, and stream gauge locations. Relational databases
include the national resources inventory, soil survey, statistical
weather data parameters, stream flow, reservoir operation data, and
agricultural census data. The internal databases allow SWAT the
ability to simulate watersheds where no gauging station data is
available. Where there are gauges, the user has the option of

214

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

importing gage data into the model or simulating data. When
modeling the Blanco River, precipitation and temperature data were
available for the area and imported into the model.

SWAT was first developed and tested in Texas. The regions used
in initial model testing include Richland and Chambers Creeks in the
Trinity region and the Rio Grande/Rio Bravo basin. Models of both
produced reasonable results when compared to measured data
(Srinivasan et al. 1998). Since its development, SWAT has been used
to model watersheds under a range of climates and land uses,
including watersheds in North Texas (Saleh et al. 2000), eastern India
(Tripathi et al. 2004), Ecuador (Hunink 2004), and Austria
(Wolkerstorfer & Strauss 2004).

Once a watershed is delineated within SWAT, the user can divide
the full drainage basin into numerous sub-basins. The number of sub¬
basins can be determined by either the user or by SWAT using
drainage patterns and tributaries in the watershed. Land use and soil
type are considered homogenous as the dominant land use and soil
type within each sub-basin for modeling purposes. Each sub-basin is
also assumed to be homogeneous in its hydrological response. SWAT
routes water, nutrients, and chemicals through these sub-basins with
input data consisting of point sources for water and nutrients.
Analytical routines model hydrology, weather, suspended sediment,
soil temperature, crop growth, nutrients, pesticides, and agricultural
management practices in each sub-basin. Routed parameters are
transformed using either a daily or monthly time step for each sub¬
basin. The results from an upstream sub-basin serve as the input for
the next sub-basin downstream.

Interfaces have been developed that allow SWAT to be used with a
Geographical Information System (GIS). One such interface is
BASINS (Better Assessment Science Integrating point and Nonpoint
Sources). BASINS was designed by the EPA as a multipurpose
environmental analysis system for use in performing watershed and
water quality studies, particularly the estimation of TMDLs (EPA
2004). With the 2001 update, BASINS added the SWAT interface

CURRAN

215

feature (EPA 2004). An advantage of using the BASINS interface is
its ability to aid in formatting input files for SWAT. These include
map layers of the land use and land cover in the watershed, digital
elevation models (OEMs), soil, slope, and watershed boundary. The
BASINS interface was used for this project.

Development of the Blanco River basin SWAT model - In order to
develop a watershed model using SWAT, critical data must first be
collected. While much can be loaded through the BASINS interface
to SWAT, a greater degree of spatial accuracy can be achieved using
outside data and importing it to the model. The model created
simulates watershed attributes based on these data. The first step in
creating a watershed involves the use of a DEM grid. The DEM is a
grid storing location coordinates with elevation. For the Blanco River
Basin SWAT model, a DEM with a resolution of 1-arc second
(approximately 30 meters or 98 feet) was downloaded from the USGS
seamless data website (USGS 2004). This DEM was used as the basis
for the modeling efforts. All of the data are projected in UTM North
America 1983, Zone 14.

Next, the watershed and stream network were delineated within
SWAT using the DEM and information on the drainage basin
downloaded from the USGS National Hydrography Database. Sub¬
basins were then defined based on where small drainages input into
the main stem of the Blanco River. The maximum size of the
threshold area for each sub-basin was defined at a resolution that
would represent the small drainages that contribute to the Blanco
River. A total of 38 sub-basins are defined within the Blanco
watershed (Figure 1). The area of each sub-basin is given in Table 1.

Layers defining the land use and soil characteristics within the
watershed were created and imported into the SWAT model. Land
use data was acquired from GIRAS Landuse/Landcover data supplied
by the USGS (USGS 2006). Soil data was acquired from the National
Resources Conservation Service STATSGO data set (NRCS: National
Resources Conservation Service 2006). Together the land use and
soil data combine to define Hydrological Response Units (HRUs)
within the watershed. Each HRU details the combination of land use

216

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

Figure 1. Map of the Blanco River watershed with the 38 sub-basins delineated.

and soil type within a sub-basin and the sub-basin area that falls into
each soil and land use category. It is through the information in the
HRUs and the stream network data that the watershed is modeled.
HRUs cross sub-basin boundaries, instead following combinations of
land use and soil type. Precipitation and flows are calculated first for
each HRU, and then for each sub-basin based on the results from the
HRUs.

The driving process in the SWAT model is the weather generator.
In the Blanco, the daily precipitation and daily minimum/maximum
temperatures records were obtained from the National Climatic Data
Center (NCDC: National Data Climate Center 2006). The
precipitation gage stations used for the Blanco River basin model
were: Blanco, San Marcos, Fischer Store, and Wimberley 1 NW.
Solar radiation, wind speed, and relative humidity were simulated by
the model due to a lack of measurement data. SWAT uses the
WEXGEN weather generator model and monthly climate statistics
calculated from long-term measured data to simulate weather data
(Sharpley & Williams 1990). Weather data for a twelve year period
was used to simulate a ten year SWAT model. The first two years of
weather data are needed in order for SWAT to simulate weather
conditions for the modeling period. In other words, the first two years
are a data preparation period so the model does not start with a dry
watershed (due to lack of precipitation data).

CURRAN

217

Table 1. Areas of the 38 sub-basin defined within the Blaneo River watershed.

Subbasin

Area

(acres)

Area

(km2)

Subbasin

Area

(acres)

Area

(km2)

6121.62

24.77

1052.37

4.26

365.84

1.48

38562.67

156.06

7092.37

28.70

3979.30

16.10

8755.65

35.43

8054.00

32.59

6352.68

25.71

6301.09

25.50

4230.60

17.12

7923.67

32.07

4809.49

19.46

9878.07

39.98

1097.74

4.44

24222.71

98.03

11572.05

46.83

17284.68

69.95

4133.41

16.73

14246.56

57.65

5542.28

22.43

9749.97

39.46

4313.55

17.46

14060.19

56.90

8355.56

33.81

6297.97

25.49

3884.78

15.72

1508.94

6.11

689.87

2.79

3939.49

15.94

5580.31

22.58

6106.94

24.71

8520.58

34.48

23242.17

94.06

789.50

3.20

4972.06

20.12

23569.98

95.38

19670.52

79.60

SWAT model calculations -Conixxmom calculation of a water mass
balance is how SWAT moves water through the model. Hydrologic
processes are calculated using surface runoff prediction and the SCS
curve number with daily rainfall records for the area. The amount of
surface runoff is calculated using a modification of the SCS curve
number method (SCS Soil Conservation Service 1972) and then the
flow is routed through the sub-basins using the Muskingham method
(Overton 1966). The Muskingham method routes water through the
tributaries and main channel while accounting for any gains or losses
to the water volume. Within the Muskingham method are parameters
to account for soil water content, surface runoff, evapotranspiration,
precipitation, water lost by seepage into the soil, and the amount of
return flow. The amount of flow moving through the sub-basins is
calculated using a daily time step.

The SCS curve number method was developed in the 1950’s to
provide a constant basis for estimating the amounts of runoff under

218

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

varying land use and soil types (Rallison & Miller 1981). It is the
choice of curve number that dictates the rate of surface runoff. For
the Blanco model, curve numbers were chosen to represent land use
and soil combination on the sub-basin scale. Within the SWAT model
are tables of curve numbers and their associated land use and soil
classes (Soil Conservation Service Engineering Division 1986),
Using the HRUs that were defined when setting up the Blanco model,
SWAT assigns curve numbers to each sub-basin. Where multiple
curve numbers could be assigned to a single sub-basin, the model
calculates a composite curve number weighted according to the
percent sub-basin area that falls within each different curve number.
Composite curve numbers are especially important for the sub-basins
containing Kyle and Wimberley. These sub-basins are combinations
of rural and urban areas, which generate significantly different runoff
rates. The user has the option to adjust curve numbers within SWAT.
It is through changes in the curve number that changes in land use,
particularly increases in impervious cover due to urbanization, are
modeled.

Once the curve number method has calculated runoff rates, the
volume of the runoff from a given precipitation event is calculated
using the Green- Ampt equation (Green & Ampt 1911). The runoff
volume is adjusted for water losses during travel to either a tributary
or the main channel. Because the climate of the Blanco watershed
includes long, hot summers, transmission losses due to the infiltration
of water into the subsurface can be significant. The quantity of runoff
after accounting for transmission losses is the volume of water that
reaches the river under different rainfall scenarios. By adjusting the
relative amounts of urban and rural surface runoff, the user can alter
the transmission losses of the precipitation as it travels to the river.
This series of changes will help predict how river flow may change
under increased urbanization.

Potential evapotranspiration (PET) is the rate at which water would
evaporate from vegetation under the condition of an unlimited water
supply. SWAT calculates PET several different ways depending on
the information supplied to the model. In the case of the Blanco River
Model, the Priestly-Taylor method was used along with temperature.

CURRAN

219

solar radiation, and relative humidity data (Priestley & Taylor 1972).
Air temperature data are supplied to the Blanco SWAT model from
gauge data. These measurements are used by SWAT to compute a
daily water temperature for a well mixed stream (Stefan &
Preud’homme 1993). Water temperature is necessary to model in-
stream biological and water quality processes.

Blanco River Basin SWAT Model Simulations

After the Blanco SWAT model was built, a number of scenarios
were modeled. Each scenario was chosen to represent a potential
change in the land use and/or climate of the watershed, with the
exception of Scenario 1, which was run to calibrate the model. Three
different land use scenarios were modeled against each of two
different precipitation regimes. The land use in Scenarios 1 and 2
reflects present day conditions for the Blanco watershed. Currently,
the watershed is approximately 2% urbanized. By adjusting the curve
numbers, the second and third land use scenarios modeled the Blanco
watershed as if it were 13% urbanized and 78% urbanized. The
estimate of 13% watershed urbanization was chosen to represent a
level that is possible within a few decades given the high rate of
population grown in central Texas. The higher urbanization rate of
78% was chosen to represent the extreme case where management
practices have allowed growth unimpeded. Both scenarios assumed
urbanization across the watershed area. The modeled precipitation
regimes were taken from existing rainfall and temperature gauge
records within the watershed. The first used data from the time frame
of 1990-2006, and the second the time frame of 1945-1960. Flows
from 1990-2006 were normal but punctuated by extreme flow events
in 1998 and 2001. 1945-1960 represents the drought of record in the
Central Texas region. Because data at Kyle date only to 1956, some
of the data was generated by the model (Nicks 1974; Richardson
1981; Richardson & Wright 1984). Each of the six SWAT model
runs created data for each of the 38 sub-basins (Table 2).

Calibration

The most effective method to determine the accuracy of the SWAT
model is by comparing the discharge data generated by the model

220

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

Table 2. Details of the different model runs.

Scenario

Land Use

Rainfall Data

Current: ~2% Urbanized

1990-2006

Current: ~2% Urbanized

1945-1960

~ 13% Urbanization

1990-2006

~ 13% Urbanization

1945-1960

~ 78% Urbanization

1990-2006

~ 78% Urbanization

1945-1960

under Scenario 1, which modeled current land use conditions with
precipitation data from 1990-2006, with discharges measured by the
USGS. Although SWAT was designed so that it does not require
calibration, the ability to compare and calibrate the Blanco model
ensures reliability of future predictions. SWAT is not a parametric
model with a formal optimization procedure. Calibration is performed
by adjusting the coefficients within the governing equations to
provide a good model fit to the watershed. In the case of Blanco
model, the curve number is adjusted. Agreement between the
modeled flows and the USGS gauge measurements need to be within
15% to give an R^>0.6 (Santhi et al. 2001). This is considered a good
fit considering the complexity of the area being modeled.

It is common practice to reduce the SCS curve numbers by 10%
from those generated by SWAT, a process that follows recom-
menddations given in Mockus 1972 (Saleh et al. 2000). The reduction
in curve numbers reflects conservation tillage practices and soil
residue cover conditions within the watershed (Santhi et al. 2001).
Having adjusted the curve numbers in the Blanco model, the modeled
flows were compared to the USGS measured flows at Wimberley
(sub-basin 28) and Kyle (sub-basin 37). In both cases, the agreement
is within acceptable parameters. The difference between modeled and
measured discharges averages 9.7% at the Kyle gauge and 3.7% at the
Wimberley gauge.

Results

Scenarios 2 through 6 predict a range of flow regimes for the
Blanco River and its tributaries. SWAT results are reported by sub-

CURRAN

221

basin, so that the flow rates in individual segments of the river can be
examined. The results are best analyzed through flow duration
curves. A flow duration curve is a cumulative frequency curve of
river stream flows. It is created by computing the percentage of time,
over a given time period, flows of specific magnitudes are equaled or
exceeded. The curve is a plot of stream flows against their respective
exceedence probabilities, and a point on the curve represents the
percentage of time the corresponding flow occurs or is exceeded for
that particular sub-basin. The flow duration curve characterizes the
ability of the watershed to provide flows over a range of magnitudes,
indicating both the likelihood of large magnitude floods and the
frequency of extremely low flows in each sub-basin.

The data in the flow duration curves was used to create
longitudinal flow profiles for each model scenario. These profiles
illustrate how flow rates increase and decrease in the Blanco River
when traveling from the headwaters to the confluence with the San
Marcos River. Using the flow duration curves, information
corresponding to the 50%, 75%, and 90% flows were determined.
The 50% flow, also called the 2-year flow, is the discharge that occurs
on average half of the period of record in each sub-basin. The 75%
flow is the discharge equaled or exceeded 75% of the time period
modeled, and it represents the lower end of flows. The 75% flow rate
is sufficiently low to characterize droughts yet maintains enough
water to show the influence of tributaries on the main channel. The
1 0% flow represents the likelihood and influence of large magnitude
floods in the watershed. These flows are rare, with a likelihood of
occurrence of only 10%. The longitudinal stream flows are given in
Figures 1 and 3. Tributary inputs were also modeled, as each tributary
is within a separate sub-basin. Because individual tributary inputs are
masked in the overall longitudinal main stem flow, tributary flows are
given in Figures 2 and 4.

Discussion

The trends in all of the simulation results show increasing flow
rates in the downstream direction. This is to be expected as the main
stem receives input from overland flow, tributaries, and possibly

222

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

Distance downstream

I Scenarios: 2% urban ♦♦♦ 13% urban BBB 78% urban A A A\

Figure 2. Results of model runs of the Blaneo River using rainfall from 1990-2006 and
land use scenarios of 2% (diamonds), 13% (squares), and 78% (triangles) urban.
These runs simulate normal precipitation conditions in the watershed. Figure 2a
shows low flows with a 75% chance of occurring, figure 2b shows flows with a 50%
chance of occurring, and figure 2c shows high flows with only a 10% chance of
occurring. Discharge rates are given in cubic feet/second (cfs) and cubic meters/
second (cms).

CURRAN

223

7 12 19 20 25 27 34 37 62 64 79 92 104 113 141 km

Distance downstream

I Scenarios: 2% urban ♦ ♦♦ 13% urban BMW 78% urban A A A|

Figure 3. Results of model runs for the tributaries to the Blanco River using rainfall from
1990-2006 and land use scenarios of 2% (diamonds), 13% (squares), and 78%
(triangles) urban. These runs simulate normal precipitation conditions in the
watershed. Figure 3a shows low flows with a 75% chance of occurring, figure 3b
shows flows with a 50% chance of occurring, and figure 3c shows high flows with
only a 10% chance of occurring. Discharge rates are given in cubic feet/second (cfs)
and cubic meters/second (cms).

224

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

Distance downstream

Scenarios: 2% urban # ^ 13% urban ■ ■ ■ 78% urban AAA

Figure 4. Results of model runs using rainfall from 1945-1960 and land use scenarios of
2% (diamonds), 13% (squares), and 78% (triangles) urban. These runs simulate
drought conditions in the watershed. Figure 4a shows low flows with a 75% chance
of occurring, figure 4b shows flows with a 50% chance of occurring, and figure 4c
shows high flows with only a 10% chance of occurring. Discharge rates are given in
cubic feet/second (cfs) and cubic meters/second (cms).

CURRAN

225

groundwater. It is the difference in the trajectories of increasing
discharge that illustrate the effects of land use management and
climate scenarios. The major tributaries to the Blanco River are the
Little Blanco River, Cypress Creek, and the Sink Creek. They are
located at 60 km (37 mi), 92 km (57 mi), and 142 km (88 mi)
downstream from the watershed headwaters respectively. Cypress
Creek joins the Blanco River at approximately the same location as
the city of Wimberley.

Assuming a precipitation regime similar to that of 1990-2006,
Figure 2 illustrates the expected flow rates throughout the length of
the main stem of the Blanco River given watershed land use that is
2%, 13%, and 78% urban. For all three land use scenarios modeled,
the flows separate between 40 and 60 km (25 and 37 mi) downstream.
This corresponds to the input of the Little Blanco River to the main
stem channel. Downstream of 60 km (37 mi), the trajectory of the
fully urbanized scenario increases dramatically.

In Scenario 5, the 13% urbanized scenario, the flows remain fairly
consistent for most of the length of the Blanco River (Fig. 2). For the
first 100 km (62 mi) downstream, there are few perturbations to the
flow rate for the low end and mid-range flows (Fig. 2a and 2b).
During flood flows, the confluence of the Little Blanco River causes a
dramatic flow increase that continues downstream (Fig. 2c). For all
flow regimes modeled with normal precipitation. Scenarios 1,3, and
5, flow increases at a downstream distance of approximately 105 km
(65 mi). This corresponds to the sub-basins downstream of
Wimberley, which may influence the rate of return flow to the main
channel.

Urbanizing 13% of the watershed does not lead to low flows that
are significantly different from those predicted under today’s land use.
However, the trends for the mid and high flow regimes predicted for
Scenario 3 mirror those predicted under Scenario 5, when the majority
of the watershed land is urban (Fig. 2). In the mid-range flows, the
flow regime increases with the same trajectory as the 78% urbanized
watershed and for the highest flow rates, the two flow regimes

226

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

coincide for the first 1 13 km (70 mi) (Fig.2b and 2c). Under all land
use scenarios, the flood flows are almost identical for the first 60 km
(37 mi), or until the confluence of the Little Blanco River. For the
Scenario 5, 78% urbanization of the watershed, the greatest change in
flow regime is at the lowest flows. The low flows are distinctly
higher when the watershed has more impervious cover (Fig. 2a).

Because each sub-basin is modeled as an individual unit, flows are
predicted for each contributing stream to the Blanco River (Fig. 3).
The majority of the inputs are minor, making the inputs from Cypress
Creek, the Little Blanco River, and Sink Creek easily discerned.
Little Blanco River has a dramatic effect under the 78% urbanization
scenario at all flows, increasing main channel flow by approximately
50%. Under today’s land use, the addition of the Little Blanco is an
important but not a critical component of Blanco River flow. When
the urbanization of the watershed is 13%, the Little Blanco input has
little effect on the flow in the main channel. Cypress Creek inputs
cause a noticeable increase in Blanco River flow only under the
scenario where 78% of the watershed is urbanized and only during
flood flows (Fig. 3c). Some tributaries are predicted to be dry under
both today’s land use condition and when 13% of the watershed is
urban. Scenarios 1 and 3. Those tributaries not predicted to be dry
contribute 0.8 cms (3 cfs) or less to the Blanco during at low flows
(Fig. 3a). This indicates that the tributaries cannot be relied upon to
keep the Blanco River flowing even under similar rainfall conditions
as during the 1990’s.

Scenarios 4, and 6 were run with rainfall data from the 1950’s to
predict flow regimes under the combination of increased urbanization
and reduced rainfall. Other than a few exceptions in the 78%
urbanized case. Scenario 6, the predicted flow regimes are all lower
than those predicted under 1990’s rainfall conditions. The difference
is most dramatic for the lowest flows (Fig. 4). For all urbanization
scenarios, the downstream flows are consistently lower until reaching
the confluence with Sink Creek. The 78% urbanized watershed has
the lowest predicted flows, with the Blanco River almost dry at many
locations. It is only at these lowest flows that the fully urbanized

CURRAN

227

scenario predicts the least flow for the channel (Fig. 4a). In contrast,
the scenarios using a 2% and 13% urban watershed area do not predict
a distinct difference in Blanco River flows. Increasing the
urbanization of the Blanco Watershed from 2% to 13% does not
significantly alter the predictions for the lowest flow rates (Fig. 4a).

Under drought conditions, the flow expected half of the time for
the Blanco River (Fig. 4b) is consistently higher when impervious
cover is increased to 78%, Scenario 6. The predications for Scenarios
2 and 4 are similar. Increasing the urban area of the watershed to 13%
deceases river flows by approximately 0.28 cms (10 cfs) from what is
predicted under current land use conditions. High flows are predicted
to be similar regardless of the rainfall scenario modeled (Fig. 2c and
4c).

The tributary influence under drought conditions is similar to that
under normal precipitation. At the lowest flows, all of the tributaries
have predicted flow rates of less than 0.28 cms (10 cfs) (Fig. 5a).
Even when compared to the mid-range flows, the majority of the
tributaries are minor contributors to the main stem. The exceptions
are under the maximum urbanization scenario, when flow rates in
both the Little Blanco River and Cypress Creek exceed 0.28 cms (10
cfs) (Fig. 5). The Little Blanco River is predicted to contribute over
0.57 cms (20 cfs), although this is a small fraction of the main stem
flow. At the highest predicted flows, contributions from the Little
Blanco and Cypress Creek are noticeable under all urbanization
scenarios (Fig. 5c).

A feature common to the different scenarios shown in Figures 2
through 5 is the order of the results. The 13% urbanized watershed
predicts the lowest flows, the 78% urbanized watershed has
consistently the highest flows, and the current land use condition of
2% urban area falls in the middle. The only exception to this trend is
the predicted lowest flows under drought conditions (Fig. 4a). The
consistency of this trend provides a strong indication of expected
changes to stream flow as urbanization progresses. As the watershed
area grows in population and infrastructure, the amount of flow
reaching the river will decrease. This is commonly due to an increase
in water usage by new residents and businesses. This water is used

228

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

Distance downstream

I Scenarios: 2% urban ### 13% urban g MB 78% urban AAA

Figure 5. Results of model runs for the tributaries to the Blanco River using rainfall from
1945-1960 and land use scenarios of 2% (diamonds), 13% (squares), and 78%
(triangles) urban. These runs simulate drought conditions in the watershed. Figure
5a shows low flows with a 75% chance of occurring, figure 5b shows flows with a
50% chance of occurring, and figure 5c shows high flows with only a 10% chance of
occurring. Discharge rates are given in cubic feet/second (cfs) and cubic meters/
second (cms).

CURRAN

229

both for household purposes and lawn watering. The accompanying
increase in impervious cover is not so great as to prevent infiltration
of rainfall into the ground. The increase in sewer connections that
could provide an increase in channel flow are not of a large enough
number to contribute a significant amount of return flow or the area
has not reached a level of urbanization where septic has been
abandoned in favor of municipal wastewater. As watershed
urbanization continues, the amount of impervious cover and hardened
land surface will reach a point where the majority of the precipitation
becomes run-off and direct recharge to the Blanco River.
Precipitation cannot infiltrate and all rainfall will run directly to the
river as overland flow. The population will have grown to the extent
that lawns are not extensive and the area is almost certainly served by
municipal wastewater services. An increase in wastewater
connections increases the amount of water the city discharges from a
treatment plant into the river. Discharges from wastewater plants and
leaky water pipes help maintain a higher baseflow in the channel
while the amount of impervious surfaces increases the speed and
amount of run-off that reaches the river during rainfall events.

Conclusions

The SWAT model has been used with the BASINS interface and
ArcView3.1 to model the Blanco River and Watershed. The
motivation of this research comes from a desire to predict future flows
in different parts of the watershed under a range of precipitation and
land use conditions. The population living in the Blanco watershed is
rapidly increasing, and is predicted to double in the coming decades.
Currently there are only two small urban areas in the watershed,
Wimberley and Kyle, and the overall watershed area is 2% urbanized.
It is expected that development growth will begin around these cities
and then expand over the watershed area.

A total of six different scenarios were modeled. All of the model
runs resulted in flow predictions for each of 38 separate sub-basins
within the larger Blanco watershed. The first scenario used rainfall
data from 1990-2006 and current land use conditions. This run was
used to calibrate the model such that predicted flow rates were within

230

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

10% of the measured flow rates at the two USGS gauging stations.
Two of the remainder model runs simulated the same precipitation but
under conditions of 13% and 78% urbanization of the watershed.
Three runs predicted watershed response during a drought, using the
1945-1960 rainfall records as input, under land use conditions that are
2%, 13%, and 78% urban.

Model runs predict an initial decrease in flows as the watershed
population grows and the urban area approaches 13% of the total land
use. Under low, mid, and high flow rates, an expansion of urban area
will first decrease flows to the river. As the amount of impervious
cover due to the infrastructure that accompanies urbanization
increases to cover the majority of the watershed area, the flows in the
river increase dramatically. Almost all of the precipitation will
become overland flow and run-off directly to the river under
conditions of extensive urbanization. When drought conditions are
simulated, there occurs an expected drop in predicted flows for all but
the most infrequent events. During a drought, the Blanco River is
predicted to maintain only limited flow with almost negligible
contribution from tributaries. These low flows continue until the
confluence of the Blanco with the Little Blanco River which adds
significant flow only during extremely high flow event. When normal
climate conditions are modeled, the contribution of the Little Blanco
River is significant to both the mid and high flow regimes. In all
scenarios, the only tributaries of significance are the Little Blanco and
Cypress Creek.

The case of 78% urbanization of the watershed area represents an
extreme, but distant possibility. In contrast, the watershed is well on
its way to becoming 13% urbanized. This research indicates flows in
the Blanco River and its tributaries will decrease consistently as
urbanization progresses in the coming decades, and some of the
tributaries will have negligible flow rates. The lowest flows are
expected over the first 60 km (37 mi) of the river, before the addition
tributary flow. This corresponds to the region of the headwaters and
spring systems, which may be of increased ecological importance.

CURRAN

231

Acknowledgements

This research was made possible by a grant from The Nature
Conservancy in Texas and the River Systems Institute at Texas State
University-San Marcos. Many thanks go to Nathanael Banda for his
patience and guidance using the model. Thanks to James Vaughn of
Texas State University-San Marcos for suggestions relative to growth
strategies of cities and Tim Bonner whose many conversations
concerning the biological nature of the Blanco watershed improved
this manuscript.

Literature Cited

(NRCS) National Resources Conservation Service. 2006. U.S. general soil map
(STATSGO), Washington, D.C.,

http://www.ncgc.nrcs.usda.gov/products/datasets/statsgo/. Accessed October 12,
2006.

Arnold, J. G. & Fohrer, N. 2005. SWAT 2000: Current capabilities and research
opportunities in applied watershed modeling. Hydrological Processes, 19:563-572.
EPA. 2004. Better Assessment Science Integrating point and Nonpoint Sources, version
3 (BASINS3.1). EPA-823-F-04-026, Washington, D.C.,

http://www.epa.gov/waterscience/basins/index.html. Accessed October 12, 2006.
Green, W. H. & G. A. Ampt. 1911. Studies on soil physics, 1. The flow of air and water
through soils, J. Agricultural Sci., 4:1 1-24.

Hunink, J. 2004. Applying the hydrological model SWAT to a watershed in the
province of Cotopaxi, Ecuador. EcoCiencia., 1-18.

Jasinski, L. E. 2006. Handbook of Texas Online.

http://www.tsha.utexas.edu/handbook/online/. Accessed January 14, 2007.

McCord, M. W. 2006. Southwest Paddler. http://southwestpaddler.com. Accessed
January 14, 2007.

Mockus, V. 1969. Hydrologic Soil-Cover Complexes. In: SCS National Engineering
Handbook, Section 4, Hydrology. USDA-Soil Conservation Service, Washington,
D.C

NCDC: National Data Climate Center. 2006. Texas climatological data. Washington,
D.C., http://www.ncdc.noaa.gov/oa/ncdc.html. Accessed October 12, 2006.

Nicks, A. D. 1974. Stochastic generation of the occurrence, pattern, and location of
maximum amount of daily rainfall. Proc. Symposium on Statistical Hydrology,
1275:154-171.

Overton, D. E. 1966. Muskingum flood routing of upland Streamflow. J. Hydrology,
4:185-200.

Priestley, C. H. B. & R. J. Taylor. 1972. On the assessment of surface heat flux and
evaporation using large-scale parameters. Mon. Weather Rev., 100:81-92.

Rallison, R. E. & N. Miller. 1981. Past, present, and future SCS runoff procedure. Pp.
353-364, in Rainfall Runoff Relationship (V. P. Singh ed.). Water Resources
Publications, Littleton, CO., xi+582 Pp.

232

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

Richardson, C. W. 1981. Stochastic simulation of daily precipitation, temperature, and
solar radiation. Water Resour. Res., 17:182-190.

Richardson, C. W. & D. A. Wright. 1984. WGEN: A Model for Generating Daily
Weather Variables, ARS-8, U. S. Dept, of Agriculture, Agricultural Research
Service, Washington, D.C.

Rodriguez, E. 2004. Austin of 2040? older, more diverse and way bigger, data predict.
The Austin American Statesman, News, Austin, TX, 1 pp.

Saleh, A. et al. 2000. Application of SWAT for the upper north Bosque River
watershed. Transactions of the AS AE, 43 : 1 077- 1 087.

Santhi, C. et al. 2001. Validation of the SWAT model on a large river basin with point
and nonpoint sources. Journal of the American Water Resources Association,
37:1169-1188.

SCS Soil Conservation Service. 1972. Section 4, Hydrology. Pp. 1-23 , in National
Engineering Handbook (SCS, ed.), U.S. Department of Agriculture, Washington,
D.C., ix+127 pp.

Sharpley, A. N. &J. R. Williams, (eds.) 1990. EPIC-erosion productivity impact
calculator, 1. model documentation. USDA Technical Bulletin 1768, 127 pp.

Short, R. A. et al. 1987. Production of Corydalus comutus (Megaloptera) in four streams
differing in size, flow, and temperature. J. N. Am. Benthol. Soc., 6:105-1 14.

Soil Conservation Service Engineering Division. 1986. Urban hydrology for small
watersheds. Technical Release 55. Washington, D.C. 164 pp.

Srinivasan, R., J. G. Arnold & C. A. Jones. 1998. Hydrologic modeling of the United
States with the Soil and Water Assessment Tool. Water Resources Development,
14:315-325.

Stefan, H. G. & E B. Preud’homme. 1993. Stream temperature estimation from air
temperature. Water Res. Bull., 29:27-45.

Tripathi, M. P. et al. 2004. Hydrological modeling of a small watershed using generated
rainfall in the soil and water assessment tool model. Hydrological Processes,
18:1811-1821.

USGS. 2006. GIRAS Landuse/Landcover data for the conterminous united states by
quadrangles at scale 1:250,000. Washington, D.C.,

http://www.epa.gov/waterscience/basins/metadata/giras.htm. Accessed October 12,
2006.

USGS. 2004. Seamless data distribution system, earth resources observation and science
(EROS). Washington, D.C. , http://seamless.usgs.gov/. Accessed October 12,2006.

White Lion & EnviroMedia. 2006. Envision central Texas.

http://www.envisioncentraltexas.org/. Accessed January 14, 2007.

Wolkerstorfer, G. & P. Strauss. 2004. Soil erosion at the mesoscale: Comparison of two
erosion models for a pre-alpine Austrian basin. lAHS, 288:339-344.

JCC at: curran@virginia.edu

TEXAS J. OF SCI. 59(3):233-240

AUGUST, 2007

COLLABORATIVE SCIENCE AND CONSERVATION IN
THE BLANCO RIVER VALLEY

Lacey E. Halstead* and Steve Jester

The Nature Conservancy, 711 Navarro, Suite 410
San Antonio, Texas 78205
*Current address:

Headwaters Coalition
Sisters of Charity of the Incarnate Word
4503 Broadway, San Antonio, Texas 78209

Abstract.-The Blanco River is 140 km (87 mi) in length and drains a watershed
of 1,067 km^ (412 mE). This central Texas river has been identified as a priority
conservation area in The Nature Conservancy’s ecoregional assessment for the
Edwards Plateau. Following its delineation as a conservation priority, a stakeholder-
driven watershed conservation plan was developed. Stakeholder involvement (46
individuals and 28 organizations) was key in shaping the Conservancy’s project plan.
Community involvement in project planning led to increased support, funding, and
partnerships for conservation and research. These results are repeatable and should
be considered as a method for leveraging applied science and conservation efforts.

The Nature Conservancy uses a rigorous planning methodology
to establish the goals and workplan for all its conservation projects.
The Nature Conservancy (the Conservancy) has developed this
approach over several decades and uses it in the United States and
internationally. The Conservancy’s planning methodology com¬
bines biological assessments with socio-cultural, political, and
economic evaluations to aid in the creation of thorough and
effective conservation plans. This methodology has been labeled
“conservation action planning,” or the “five-s framework.” The
term “five-s” stands for the five main components of the method,
outlined below.

(1) Systems: Selection of the species, communities and
ecological systems that will be conservation priorities, and
an assessment of their long-term viability.

(2) Stresses: Assessment of harmful biological impacts seen in
priority systems now or expected to arise within the next 10
years.

234

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

(3) Sources of Stress: Evaluation of the activities causing each
stress, projected over 10 years.

(4) Strategies: An assessment of the socio-cultural, political,
and economic context within which conservation must
occur and the setting of strategies that will abate sources of
stress and ensure long-term viability of systems.

(5) Success: Establishment of metrics by which to track
conservation progress.

This methodology is science-based, but also designed to enable
action in the face of scientific uncertainty. Therefore, to work well,
conservation action planning must include local experts who can fill
in data gaps with their observations and hypotheses. The traditional
five-s planning process involves Conservancy staff, natural
resource professionals, other non-profits, and academics. However,
in large landscapes, particularly those with considerable private
land, the traditional process excludes many key stakeholders.
Therefore, the Texas chapter of The Nature Conservancy uses a
modified five-s process to plan with large community groups.
Inviting community members to help shape projects has better
engaged and excited stakeholders of all sorts, and contributed to
numerous collateral benefits. Among these are increased

acquisition of local knowledge, synergy among scientists and
agency land managers, greater donor support for science and
conservation work, and more positive community involvement.
These benefits are repeatable, and such community-based planning
can help advance science and conservation.

This paper highlights aspects of the collaborative planning
process as it was used in the Blanco River watershed, but focuses
primarily on results and benefits. Resources are available for those
wishing to learn how to conduct a traditional (The Nature
Conservancy 2000) or collaborative (The Nature Conservancy
2003; Halstead & Jester 2004) five-S planning process.

HALSTEAD & JESTER

235

Planning Process

The Blanco River conservation area spans 114,063 ha (281,846
ac) in the Edwards Plateau ecoregion of central Texas (Figure 1).
The river flows through Blanco and Hays counties and ends at the
San Marcos River, inside the city of San Marcos. For years,
watershed residents have been collaborating in a more or less ad-
hoc fashion to retain their natural resources, working mainly within
small grassroots organizations. The Conservancy initiated a con¬
servation project in the watershed in n 2002 to expand upon this
work. [

To capitalize on ongoing local efforts, the Conservancy chose to
engage stakeholders in the project from the Initial planning stage.
The first step in this engagement was to initiate a dialogue with
community members. Conservancy staff spent about three months
building relationships with local residents, meeting key stake¬
holders, and learning community dynamics. This vital step
involved numerous one-on-one conversations, extensive requests
for contact referrals, and a great deal of listening. This was a key
component to building a planning team, and to ensuring adequate
participation in collaborative planning.

The second step was to invite about 50 stakeholder
representatives to two planning meetings. Both meetings included
technical experts (traditional planning partners) and stakeholders
from across the watershed. Groups and organizations invited were:
small-tract landowners, large-tract landowners, Blanco Chamber of
Commerce, Blanco-Pedemales Groundwater Conservation District,
Cypress Creek Conservation Association, Texas Council on
Environmental Quality, Edwards Aquifer Research and Data
Center, Texas Farm Bureau, Guadalupe Blanco River Authority,
Texas House of Representatives, Guadalupe-Bianco River Trust,
Texas Parks and Wildlife Department, Hays County Grants
Administration, Texas State University San Marcos, University of
Texas, Hays Trinity Groundwater Conservation District, The
Nature Conservancy, Hill Country Conservancy, U.S. Fish and
Wildlife Service, Institute for Sustainable Water Resources, Blanco

236

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

Austin

New Braunfels

San Antonio,

i; (IlOEO®£)(a0!^ f

20 Kilometers

Figure 1. Blanco River Conservation Area.

County Commissioners Court, Wimberley Chamber of Commerce,
Hays County Commissioners Court, Wimberley Valley Watershed
Association, Mayor of Blanco, Mayor of Wimberley, and
Wimberley Water Supply. Invited stakeholders included people
considered predisposed to the Conservancy’s methods and mission,
and those likely to be opposed to the organization. Oral invitations
were extended to most participants by the project director, followed
by letters of invitation. Non-respondents received telephone calls
encouraging them to attend. This three-part effort garnered a few
additions to the original list, for a total of 46 attendees in two

HALSTEAD & JESTER

237

workshops. Stakeholders were invited to both workshops, and
about one-third attended both.

The third step in the process was the planning itself. Four
facilitators took the diverse community planning team through the
five-s process. The first meeting, which was more heavily attended
by laypeople, involved each step of the process except the viability
assessment. The second meeting built upon the first, with many of
the same participants, although attendance was weighted toward
scientists and resource managers. This group quality-checked the
work to date, and added the viability assessment. Both meetings
lasted just under two days, during which all participants shared
equally the task of creating a conservation plan.

After these meetings, Conservancy staff synthesized the
planning team’s work into a draft conservation plan, which was
reviewed by all participants and other local residents before
publication and adoption. This plan, once approved, became the
five-year workplan for the Conservancy’s Blanco River Project
(The Nature Conservancy 2004). The process, from initiation to
plan completion, took slightly less than eight months.

Results and Discussion

The resulting conservation plan included considerable detail
about systems, viability, stresses and sources of stress, as well as
strategy implementation (The Nature Conservancy 2004a). A
summary of the major strategic initiatives is given here to
demonstrate some of the innovations that came from the
collaborative process. The first two initiatives represent fairly
typical Conservancy strategies; while the last two are novel ways to
help accomplish the Conservancy’s mission and meet the
community’s needs. These two initiatives would not have been
adopted, or perhaps even considered, in a traditional planning
process.

(1) Sustainable water w^^.-Ecologically sustainable use of
groundwater and surface water in the Blanco River basin is
the norm.

238

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

(2) Sustainable land management. -VnvditQ lands management
is conducted in such a way as to allow maintenance or
recovery of diverse native plant and animal communities in
riparian forests, floodplains, canyons, and uplands.

(3) Environmentally sensitive development.— Eoo\og\Q2i\\y sus¬
tainable building is the norm in the conservation area,
especially practices that minimize habitat loss, habitat
fragmentation, impervious cover, and unsustainable water
use.

(4) Natural and cultural heritage. Blanco River conserva¬
tion area continues to be known as a place with a rich
natural and cultural heritage and a strong conservation ethic,
which the Conservancy helps promote.

Most natural resource plans are developed by professionals and
at some point provided to the public or a select group of
stakeholders for review. The Blanco River watershed planning
effort involved stakeholders throughout the process, building the
plan rather than reacting to it. This increased the knowledge base
available to the Conservancy, created a sense of ownership among
participants, and led to greater support for the project long-term.
Also, by using a facilitated, collaborative process to step
participants through the detailed analyses inherent in conservation
action planning, the Conservancy was able to help participants with
varied backgrounds gain a deeper understanding of biological,
social, political, and economic issues influencing conservation in
the watershed. Local landowners and suburbanites worked beside
research scientists to select the conservation priorities (systems) in
the watershed. They collectively analyzed the viability of, and
threats to, these priority systems, sharing data and anecdotal
knowledge. Academic scientists and agency staff brainstormed
conservation strategies with developers, ranchers, and part-time
homeowners. This diverse team helped to build a conservation plan
that was more comprehensive, more balanced, and more widely
supported than a traditionally constructed plan would have been.

HALSTEAD & JESTER

239

In addition to more effective conservation strategies, working
through a collaborative planning process with communities in the
watershed helped advance science and on-the-ground action. Even
before the first meeting, $250,000 had been pledged to support
science and conservation. The funding came from donors who
were excited about supporting a holistic conservation project that
included stakeholders in a meaningful way, and that led to on-the-
ground results in a short timeframe. The funded research has led to
a detailed assessment of the biological diversity supported by the
river as well as the critical dynamics (flow volume and seasonal
variation) that make the Blanco River ecologically functional.
Also, the investment in the science needs of this project garnered a
key science partner (Texas State University) and thus greater
credibility within the community. The plan and resultant research
continue to help build support from funders, partners, and members
of the community. For example, another by-product of the
planning collaboration was a well-received riparian management
guide produced by the Conservancy, Natural Resources Conserva¬
tion Service, Guadalupe-Bianco River Authority, and Nueces River
Authority.

This process provided other benefits that will be ongoing: within
four months of the first community meeting, 100 landowners had
been exposed to the project, and the open process improved the
Conservancy’s reputation among these and other stakeholders. This
has led to increased access on private lands for data gathering, and
new land stewardship partnerships.

Some products of collaborative planning are less tangible.
Participants learned about each other’s viewpoints, values and
needs and came to know each other as individuals. As a result, they
were more open to hearing from and accommodating people whom
they had traditionally opposed. This shift led to better cooperation
and less animosity over controversial issues. For example, many
residents started this process convinced that developers added to
urban sprawl because they “didn’t care” about the environment, and
were not interested in ecologically compatible development. After

240

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 3, 2007

working through the assessments involved in five-s, residents
learned from developers in attendanee that zoning laws and
economics often prevented construction of ecologically compatible
subdivisions. The planning team then recommended two
collaborative solutions: (1) a study to demonstrate the economic
feasibility of ecologically compatible development and (2)
encouraging the county to adopt a voluntary “conservation
development” option that would incentivize rather than penalize
clustered development.

Collaborative conservation planning served as a great first step
in this complex, community-based watershed project. Proper
functioning of the entire watershed is critical to conserving
biodiversity here, and the land is almost entirely privately owned.
The number of landowners making individual management
decisions is huge, and their motivations are diverse. Utilizing a
process where divergent views can be incorporated with data and
expert opinion is critical for organizations seeking to contribute
positively to conservation and research in similar situations.

Literature Cited

Halstead, Lacey & Steve Jester. 2004. Lessons learned in community-based
planning: Using the Nature Conservancy’s participatory conservation planning
process. The Nature Conservancy, San Antonio, Texas. Available at
www.conserveonline.org

The Nature Conservancy. 2000. The five-s framework for site conservation: a
practitioner’s handbook for site conservation planning and measuring
conservation success. Vol. I. The Nature Conservancy, Arlington, Virginia, 58

pp.

The Nature Conservancy. 2003. Participatory conservation planning manual. The
Nature Conservancy, Lore Lindu field office, Sulawesi Tengah, Indonesia, 33

pp.

The Nature Conservancy. 2004. A biodiversity and conservation assessment of the
Edwards Plateau ecoregion. Edwards Plateau ecoregional planning team. The
Nature Conservancy, San Antonio, Texas, 121 pp.

The Nature Conservancy. 2004a. Conservation area plan for the Blanco River. The
Nature Conservancy, San Antonio, Texas, 50 pp.

LHD at: lacey.halstead@amormeus.org

THE TEXAS ACADEMY OF SCIENCE, 2007-2008

OFFICERS

President
President Elect
Vice-President.

Immediate Past President.

Executive Secretary:

Corresponding Secretary:

Managing Editor:

Manuscript Editor:

Treasurer:

AAAS Council Representative:

International Coordinator:

DIRECTORS

2005 Jerry L. Cook, Sam Houston State University
Flo Oxley, Lady Bird Johnson Wildflower Center

2006 Herbert D. Grover, Hardin-Simmons University
Gary P. Garrett, Texas Parks and Wildlife Department

2007 Renard L. Thomas, Texas Southern University
Bob Murphy, Texas Parks and Wildlife Department

SECTIONAL CHAIRPERSONS

Anthropology: Roy B. Brown, Institute Nacional de Antropologia y Historia

Botany: Alan W. Lievens, Texas Lutheran University

Cell and Molecular Biology: Jon B. Scales, Midwestern State University

Chemistry and Biochemistry: Benny E. Amey, Jr., Sam Houston State University

Computer Science: Laura J. Baker, St. Edward’s University

Conservation Ecology: Cathy Early, University of Mary Hardin Baylor

Environmental Science: Kenneth R. Summy, University of Texas-Pan American

Freshwater and Marine Sciences: Brian W. Brooks, Baylor University

Geosciences: Carol Thompson, Tarleton State University

Mathematics: Sandra Luna McCune, Stephen F. Austin State University

Physics: David Bixler, Angelo State University

Science Education: Kaycie Sullivan, Texas Tech Campus at Junction

Systematics and Evolutionary Biology: Allan W. Hook, St. Edward’s University

Terrestrial Ecology and Management: Christopher M. Ritzi, Sul Ross State University

COUNSELORS

Collegiate Academy: William J. Quinn, St. Edward's University
Junior Academy :VincQ Schielack, Texas A&M University

Hudson R. DeYoe, University of Texas-Pan American

Raymond C. Mathews, Jr., Texas Water Dev. Board

William J. Quinn, St. Edward’s University

David S. Marsh, Angelo State University

Fred Stevens, Sehreiner University

Cindy Contreras, Texas Parks and Wildlife Department

Ned E. Strenth, Angelo State University

Frederick B. Stangl, Jr., Midwestern State University

John A. Ward, Brooke Army Medical Center

James W. Westgate, Lamar University

Armando J. Contreras, Universidad Autonoma de N.L.

THE TEXAS JOURNAL OF SCIENCE PERIODICALS

Texas Academy of Science

CMB 6252

Schreiner University

Kerrville, Texas 78028-5697

)

TEXAS

THE

JOURNAL

SCIENCE

Volume 59
Number 4
November 2007

GENERAL INFORMATION

MEMBERSHIP. -Any person or member of any group engaged in
scientific work or interested in the promotion of science is eligible for I
membership in The Texas Academy of Science. For more informa¬
tion regarding membership, student awards, section chairs and vice-
chairs, the annual March meeting and author instructions, please ac- ;
cess the Academy's homepage at: '

www.texasacademyofscience.org S

Dues for regular members are $30.00 annually; supporting mem- \
bers, $60.00; sustaining members, $100.00; patron members, $150.00; ’

associate (student) members, $15.00; family members, $35.00; affili- ;
ate members, $5.00; emeritus members, $10.00; corporate members,
$250.00 annually. Library subscription rate is $50.00 annually.

Dr. Fred Stevens, Executive Secretary
The Texas Academy of Science
CMB 6252
Schreiner University
Kerrville, Texas 78028-5697
E-mail: FStevens@schreiner.edu

The known office of publication for The Texas Journal of Science is the Department of Biology,
Angelo State University, San Angelo, Texas 76909; Dr. Ned E. Strenth, Managing Editor.

THE TEXAS JOURNAL OF SCIENCE

Volume 59, No. 4

November, 2007

CONTENTS

Foraging Preferences by American Beavers, Castor canadensis
(Rodentia: Castoridae) on Central Texas Rivers.

By John T. Baccus, Mark A. Kainer and Michael F. Small . 243

Effects of Permanent Water on Home Ranges and Movements of Adult Male
White-Tailed Deer in Southern Texas.

By Stephen L. Webb, David G. Hewitt and Mickey W. Hellickson . 261

Spatial and Temporal Expression of Syndecan - 2 (Fibroglycan) in
Chick Heart Development.

By Ashley Gordon and J. Kevin Langford . 277

A Key to the Common Seed Shrimp (Crustacea: Ostracoda) of
the Playa Lakes of the Llano Estacado Region of Northwestern Texas.

By Francis R. Horne and Ned E. Strenth . 29 1

General Notes

A Formal Synthesis of (+)-Muricatacin from L-Tartaric Acid.

By Jack M. Southard & Nicole L. Sears . 301

Re-Occurrence of the Tropical Green Macroalga, Penicillus capitatus
(Chlorophyta: Bryopsidales), in the Lower Laguna Madre of South Texas.

By Joseph L. Kowalski, Donald L. Hockaday, Gilbert H. Boza, Jr.,

and Hudson R. DeYoe . 305

Reproduction in the Redback Coffee Snake, Ninia sebae
(Serpentes: Colubridae), from Southern Mexico and Central America.

By Stephen R. Goldberg . 3 1 1

Possible Vertebrate Burrows from the Miocene Fleming Formation
near Huntsville, Walker County, Texas.

By Thomas A. Stidham . 317

A Noteworthy Infection of Clinostomum complanatum (Digenea: Clinostomidae)
in a Cave Salamander, Eurycea lucifuga (Caudata: Plethodontidae), from
Northcentral Tennessee.

By Chris T. McAllister, Charles R. Bursey, Matthew L. Niemiller

and Brian T. Miller . . . . . . . . . . . . . 321

Index to Volume 59 (Subject, Authors & Reviewers) . . 327

Recognition of Member Support . . . . 333

Membership Application . . . . . 334

Postal Notice......... . . . . . . . . . 335

THE TEXAS JOURNAL OF SCIENCE
EDITORIAL STAFF

Managing Editor:

Ned E. Strenth, Angelo State University
Manuseript Editor:

Frederiek B. Stangl, Jr., Midwestern State University
Associate Editor for Botany:

Janis K. Bush, The University of Texas at San Antonio
Associate Editor for Chemistry:

John R. Villarreal, The University of Texas-Pan American
Associate Editor for Computer Science:

Nelson Passos, Midwestern State University
Associate Editor for Environmental Science:

Thomas LaPoint, University of North Texas
Associate Editor for Geology:

Ernest L. Lundelius, University of Texas at Austin
Associate Editor for Mathematics and Statistics:

E. Donice McCune, Stephen F. Austin State University
Associate Editor for Physics:

Charles W. Myles, Texas Tech University

Manuscripts intended for publication in the Journal should be submitted in
TRIPLICATE to:

Dr. Frederick B. Stangl, Jr.

TJS Manuscript Editor
Department of Biology
Midwestern State University
Wichita Falls, Texas 76308
frederick.stangl@mwsu.edu

www.texasacademyofscience.org

AFFILIATED ORGANIZATIONS
American Association for the Advancement of Science,

TEXAS J. OF SCI. 59(4):243-260

NOVEMBER, 2007

FORAGING PREFERENCES BY AMERICAN BEAVERS,
CASTOR CANADENSIS (RODENTIA: CASTORIDAE) ON
CENTRAE TEXAS RIVERS

John T. Baccus, Mark A. Kainer and Michael F. Small

Wildlife Ecology Program, Department of Biology
Texas State University, San Marcos, Texas 78666

Ahstract.-The primary objective of this study was to assess forage preference of
woody vegetation by American beavers {Castor canadensis) on rivers in central
Texas. Foraging data collected for beavers on three primary rivers in central Texas
during 1990 and 1991 provided information on diet, food preferences and size class
selectivity of woody plants. White mulberry {Morns alba) and boxelder {Acer
negundo) had the greatest use on all rivers. Beavers preferred eastern cottonwood
{Populus deltoides), white mulberry, Mexican buckeye {Ungnadia speciosa), black
willow {Salix nigra), and bur oak {Quercus macrocarpa) on all rivers and avoided
Chinaberry {{Melia azedarach), hackberry {Celtis sp.), cedar elm {Ulmus crassifolia),
and pecan {Carya illinoensis). One to 5 cm stems were the most preferred size class
for all woody vegetation. The ecological implications of selective foraging by
beavers on availability and structure of woody vegetation are substantial on rivers
with narrow riparian zones in central Texas.

American beaver {Castor canadensis Kuhl) occurs naturally in
nearly every available aquatic habitat in North America, with the
exceptions of the arctic tundra, high montane environs, south¬
western deserts, and peninsular Florida (Hall 1981; Jenkins &
Busher 1979). Beavers have been categorized by diet as
vegetarians, choosy generalists, opportunistic herbivores, or
obligate vegetarians (Johnson 1927; Harper 1969; Jenkins 1978,
1981; Busher 1996; Baker and Hill 2003; Schmidly 2004). They
consume a variety of plant parts (bark, shoots, foliage, nuts, and
roots) but preferentially select foods based on species, size class,
distance to source, and nutritional quality (Jenkins 1975, 1980,
1981; Pinkowski 1983; Belovsky 1984; McGinley & Whitham
1985; Fryxell & Doucet 1991; Fryxell 1992; Doucet & Fryxell
1993). Beavers prefer a greater variety of herbaceous vegetation in
their diet than woody plants during all seasons (Henderson 1960;
Jenkins 1979, 1981; Svendsen 1980; Roberts & Amer 1984), but
woody plants are the limiting dietary factor for their presence on

244

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

first through third order streams with moderate to low flow rates
(Slough & Sadlier 1977).

Woody plants in beaver diets vary geographieally relative to
plant availability. In northeastern North America, beavers preferred
aspen {Populus tremuloides), and Johnson (1927) suggested an
evolutionary relationship existed between the geographic origin of
C. canadensis and Populus sp. Beavers frequently fed on yellow
poplar (Liriodendron tulipifera), American hornbeam (Carpinus
carolinia), red maple (Acer rubrum), and sugar maple (A.
saccharinum) in the Midwest (Svendsen 1980). Willow (Salix sp),
poplar (Populus balsamifera), and alder (Alnus crispa) were
common foods in the diet of beavers in the Canadian Northwest
Territories (Aleksiuk 1970).

Beavers rely on woody vegetation for the majority of their diet
from October through April (Svendsen 1980; Hill 1982; Roberts &
Amer 1984). Denney (1952) ranked preference of woody plants by
beavers in North America as aspen, willow, cottonwood (Populus
deltoides), and alder. However, caution is warranted when making
conclusions relating to preference; extensive use does not imply
preference when a greater diversity of trees produces complex
patterns of food selection (Jenkins 1981). Little is known about
beaver food preference for most regions of its range, especially
southern populations, with only relative food preferences
documented for a few areas of northeastern and western North
America (Jenkins 1981). Beavers select forage resources at
different levels of resolution, but little information is available
concerning species preference, selective foraging effects on
structure and composition of riparian vegetation, and the
mechanisms involved (Jenkins 1981).

As a species portraying aspects of central place foraging theory,
beavers exert choices upon a variety of prey sizes. Studies on prey
size selection by foraging beavers in various habitats suggested
selection for trees of smaller diameters with increasing distance

BACCUS, KAINER & SMALL

245

from water (Jenkins 1980; Pinkowski 1983; Belovsky 1984;
McGinley & Whitham 1985; Donkor & Fryxell 1999, 2000; Gallant
et al. 2004). Donkor and Fryxell (1999, 2000) found fewer stems
selected by beavers as a function of distance from water, and
Fryxell and Doucet (1993) demonstrated size selection for aspen
and alder varied as availability changed. Basey et al. (1988)
documented that the greater selectivity for prey sizes displayed by
beavers was consistent with central place foraging theory (Fryxell
& Doucet 1991).

There is a paucity of information on foraging and food habits for
beavers in Texas. Only general information on the variety of foods
consumed by beavers is available (Schmidly 2004), and this
information is restricted to winter foods of beavers in central Texas.
The objectives of this study were to assess forage preference of
woody vegetation by beavers, determine size class preference, and
investigate niche breadth with respect to forage resources on the
San Marcos, Blanco, and San Gabriel rivers in central Texas.

Methods

Study area.-The study was conducted in the riparian zone along
8 km stretches used as foraging sites by beavers on the San Marcos
(Hays County), Blanco (Hays County), and San Gabriel
(Williamson County) rivers in central Texas (Fig. 1). These rivers
originate near the southeastern perimeter of the Edwards Plateau
Ecological Region and flow southeasterly across the Balcones
Escarpment into the Blackland Prairie Ecological Region (Gould et
al. 1960). Riparian zones of these rivers vary in width based on the
extent of adjacent land-use practices.

The San Marcos River is a spring-fed, perennial stream with a
relatively constant flow and temperature (Tupa & Davis 1976,
Groeger et al. 1997). The study area extended from Spring Lake in
San Marcos (29^ 52’ 23” N, 9T 56’ 06” W) to the confluence of
the river with the Blanco River. Along this stretch, the floodplain
was slightly rolling with relatively steep river banks. The river was

246

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

Figure 1. Locations of foraging stretches used by American beavers on the San Marcos,
Blanco and San Gabriel rivers in central Texas.

moderately used for recreation throughout the year, but the stretch
within the city was heavily used, especially in spring and summer.
Outside the city, land-use adjacent to the river was agricultural.
Woody species of the riparian zone included bald cypress
(Taxodium distichum), black willow {Salix nigra), American and
cedar elm {Ulmus americana, U. crassifolia), pecan {Cary a

BACCUS, KAINER & SMALL

247

illinoensis), chinaberry (Melia azedarach), hackberry {Celtis sp.),
eastern cottonwood {Populus deltoides), boxelder {Acer negundo),
sycamore (Platanus occidentalis), privet (Ligustrum sp.), elderberry
{Sambucus canadensis), rough-leaf dogwood {Cornus drummondii),
Mexican buckeye (Ungnadia speciosa), red buckeye (Aes cuius
pavia), anacua (Ehretia anacua), bur oak (Quercus macrocarpa),
white and red mulberry {Morus alba, M. rubra), honey mesquite
{Prosopis glandulosa), and Texas persimmon (Diospyros texana).

The Blanco River site extended from the Interstate Highway-35
bridge (29° 54’ 09” N, 97° 53’ 52” W) to the confluence with the
San Marcos River. Environs on this stretch were a combination of
steep cliffs on meandering curves with several backwater areas and
relatively straight segments with well-developed floodplains. The
river was perennial, but base flow was much lower and fluctuates
more than the San Marcos or San Gabriel rivers. Woody vegetation
was similar to the San Marcos River, however, areas adjacent to the
Blanco River had a greater abundance of sycamore, Texas ash
{Fraxinus texensis), and black willow trees and fewer understory
species. Adjacent land-use was largely agricultural with remnants
of gravel mining. Recreational use was minimal.

The San Gabriel River site began at the Texas Highway 95
bridge (30° 38’ 39” N, 97° 26’ 21” W) and continued 8 km
downstream. Topography along the river was similar to the Blanco
River. Land-use was largely agricultural. Recreational use was
minimal. The woody plant community was similar to the San
Marcos and Blanco rivers with the addition of osage-orange
{Madura pomifera), western soapberry {Sapindus saponaria), live
oak {Quercus virginiana), and Texas sophora {Sophora affinis).

Data collection and analysis -DiumdiX and nocturnal canoe
surveys were conducted to identify active foraging areas of beavers
on each river in spring 1990 and 1991. In a trial study in 1988 and
1989, beavers foraged up to 110 m from the river bank, but 95% of
18 woody species used by beavers occurred within a 5-m riparian
zone. Based on the trial study, all tree and shrub trunks with a stem

248

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

diameter >1 cm within a 5-m belt transect were identified for each
forage stretch on each river. Each trunk was examined for beaver
use and classified into six discreet size classes (1-5, 6-9, 10-13, 14-
17, 18-21, and >21 cm). Individual main branches of plants with
multiple trunks were designated as separate individuals for
purposes of beaver usage and availability counts. Stem diameter
was measured at 50 cm above ground for comparability between
felled and girdled stems and compensation for interspecific
variation in basal trunk spread. Use was defined as a tree or shrub
felled from girdling or >25 % girdling of the trunk. Foraging on
trees by animals other than beavers was observed. Therefore, data
for woody plants with main stem diameters <1 cm were excluded
from analyses because nutria {Myocastor coypus) and swamp
rabbits {Sylvilagus aquaticus), known to occur at study sites, forage
on small woody saplings (Blair & Lauglinais 1960).

Woody resource use and preferences were analyzed by four
calculations. Hurlberf s standardized niche breadth index (Hurlbert
1978; Krebs 1999) was used to evaluate woody resource usage.
This method allowed scaling of resources based on their availability
when evaluating niche breadth. Niche breadth was calculated for
the three rivers each year to determine whether plant use by beavers
varied temporally. Usage and availability data for 1990 and 1991
for each river were combined and niche breadth calculated to detect
any spatial variation among rivers. Woody plants were designated
as frequently used when quantities of the total number of plants for
each river exceeded 10% (D. Johnson, pers. comm.).

Variance in patterns of relative preference for woody plants and
dependence within size classes for each species among rivers were
evaluated by a 3 x 16 contingency table. The proportions of used
to available individuals for each plant species combined for both
years on each river were transformed to count data (Zar 1984).
Z(x^) was used for each row (representing species relative use for
all rivers) to determine species contribution to observed differences
in relative preference patterns among rivers. The null hypothesis

BACCUS, KAINER & SMALL

249

that patterns in relative preferences for size classes were
independent of rivers was tested with a ^ contingency analysis.

A rank preference index (Johnson 1980; Krebs 1999) was used
to determine relative preferences for woody plant species within
each river stretch because the inclusion or exclusion of common but
seldom-used species in preference indices may mask preference or
avoidance of species (Krebs 1999). A food type was considered
preferred (selected, favored) if it constituted a significantly larger
fraction of the foraged items than of an unbiased sample of the
various food types available (Jenkins 1981). Values <0 indicated
preference and values >0 indicated avoidance. For this study,
beaver colonies represented individual subjects (D. Johnson, pers.
comm.) and resource states were species of woody plants used by
beavers. A mean difference in ranks for each resource was also
applied in assessing preference. A lower value for the mean
difference in rank indicated selection for the resource; whereas, a
higher positive mean difference indicated avoidance.

The Waller-Duncan procedure (Johnson 1980) was used to test
whether beavers equally preferred all woody species on each river.
An F-statistic was calculated by multiple comparisons between
plant species used. The relative preference for the six size classes
of stems within each species was determined by Ivlev's Electivity
Index (Krebs 1999). Ivlev's electivity values >0 indicated relative
preference for the resource.

Results

Beavers foraged on woody plants in 40 river stretches of
variable spatial and temporal lengths. Lengths of foraging stretches
did not differ from 1990 to 1991 (San Marcos: x = 167.54 m, SE =
11.296, min = 117 m, max = 164 m, 4 = -1.58, P = 0.1741; Blanco:
X == 157.13 m, SE = 9.7399, min =112 m, max = 147 m, 4 = -0.57,
P = 0.5904; and San Gabriel: F = 158.75 m, SE = 11.582, min =
112 m, max = 151 m, 4 = -0.02, P = 0.9865) or among foraging
stretches on the three rivers (F2, 37 = 0.27, P = 0.7646).

250

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

Discreet beaver colonies identified by forage areas were
separated by >200 m on the three rivers. Percent usage of foraging
stretches were similar for the three rivers for 1990 and 1991 (San
Marcos: 13.6% for both years, Blanco: 13.5% and 15.9%,
respectively, and San Gabriel: 11.8%) and 11.9%o, respectively).
Additional forage stretches located in 1991 indicated beaver use of
relatively marginal habitats and establishment of new colonies.

Beavers foraged on 6,532 (24.9%o) of 26,242 available woody
trunks of 18 woody species (Table 1). Niche breadth among rivers
was similar in two respects. Hurlbert's niche breadth values varied
temporally among rivers, with the San Gabriel River having the
lowest and highest Hurlbert's niche values (0.813 in 1990, 0.902 in
1991, and 0.862 combined). The San Marcos River had the highest
overall Hurlbert's niche value (0.833 in 1990, 0.895 in 1991, and
0.881 combined); whereas, the Blanco River had the lowest
combined value (0.814 in 1990, 0.863 in 1991, and 0.835
combined). Overall, beavers foraged on similar woody plants along
all three rivers. Also, the equitability of woody plant use between
years and among rivers were similar, with >10%o of total individuals
of four species used. The four species frequently used varied
among rivers; however, these species were the same for each river
over both years.

The overall pattern for relative species preference differed
among rivers (/^j 3^ = 84.34, P <0.001). Rough-leaf dogwood
accounted for the greatest deviation from expected frequencies =
28.7). Additionally, bur oak, Chinaberry, eastern cottonwood, and
privet contributed to foraging heterogeneity. Boxelder and white
mulberry were the most frequently used woody plants on all rivers.
Rough-leaf dogwood, black willow, boxelder, privet, and white
mulberry comprised 77.4%o of woody plant usage on the San
Marcos River. Black willow, boxelder, sycamore, and white
mulberry represented 79.6%o of woody plant usage on the Blanco
River. Black willow, boxelder, Texas ash, and white mulberry
accounted for 89.3%o of woody plant usage on the San Gabriel
River.

BACCUS, KAINER & SMALL

C73

D-

O-

“t

•-t

<T>

CfO

<■

o’

ty>

u»

ty>

tot

-o

45-

-o

45-

Cft

Cfi

1 — 1

45-

-o

ty>

45-

t/>

t.n

■o

45.

t/1

u->

t/1

-o

45-

t/»

45.

45-

1— >

-o

t.^

ty«

t/»

45-

.t-

45-

45.

45-

-o

-p

-o

-.O

-o

1—,

L/«

■45-

45-

4:-

45-

t.n

45-

tyl

-o

tyi

■-o

-o

--J

45-

t/l

45.

-o

■o

45-

ty>

t./<

•0

1— >

45-

1— •

>— >

-o

t/1

ty>

tvj

t/«

H- *

t/>

-o

45-

Cfi

45-

-o

t./1

tOl

45-

on
2-
3 CD
O-

X <

3 SI
X c?

251

Table 1. Frequency of individual stems foraged by American beavers and woody resource availability by species within forage stretches
along the San Marcos, Blanco and San Gabriel rivers in 1990 and 1991. The selectivity index is the proportion of stems selected
based on usage and availability of each woody species.

San Marcos Blanco San Gabriel

252

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

By combining proportions of frequently used species for all
rivers over both years, boxelder (25%), black willow (14.1%), and
white mulberry (15%) accounted for 54.1% of all woody species
used. Beavers used blaek willow only on the Blanco and San
Gabriel rivers. Rough-leaf dogwood and privet had frequent use on
the San Marcos River (18% and 21%, respectively, by year), slight
use on the Blanco River (1% and 2%, respectively, by year), and no
use on the San Gabriel River. Texas ash was frequently used on the
San Gabriel River (24%) but had minimal usage on the Blanco
(6%) and San Marcos (0.2%) rivers. Similarly, sycamore had
higher usage on the Blanco River (13.7%) compared to the San
Marcos (1.3%) and San Gabriel (3.9%) rivers.

Relative species preference by beavers differed on all rivers (^4,9
= 4.400, P < 0.025, San Marcos River; ^5, 9 = 3.550, P < 0.025,
Blanco River; ^3,9 = 4.33 1, P < 0.025, San Gabriel River). Data for
both years indicated beavers preferred eastern cottonwood, white
mulberry, Mexican buckeye, black willow, and bur oak on all rivers
(Table 2) and avoided Chinaberry, hackberry, pecan, cedar elm, and
boxelder on all rivers (Table 3).

Relative preferences for size classes among rivers did not differ
10 = 7.12, P > 0.05) with the 14 to 17-cm and 18 to 21-cm size
elasses showing the greatest differences. Beeause patterns of
relative preference for size classes of woody plants were
independent of rivers, data for all rivers over both years were
combined. Size class was correlated to an index of electivity, with
the 1 to 5-cm size class the most preferred size class for all woody
species except for rough-leaf dogwood, Mexican buckeye, and bald
cypress (r^ = 0.998, Fj, 3 = 363.72, P > 0.001) (Fig. 2, Table 4).
Beavers slightly preferred the 6 to 9-cm size class of rough-leaf
dogwood over the 1 to 5-cm size class (Ivlev's electivity values of
0.028 and -0.005, respectively). Beavers showed a relative
preference for the > 21-em size class for bald cypress and
cottonwood.

BACCUS, KAINER & SMALL

H O dO W > O OC

pr o P 3 H) p

-3 ! 3^

—

L/v

■0

•0

O CO W

O
X
n

3 a-

n X

O 2 ^
^ o

c/5

O Q. o

p ^

o ft*

^ 3

—

■4^

-o

p p p p ^ ^ u>

u) u) bo o uj ov !p.

'-J vO UJ K) tyi

ui u> to UJ Ui oo

c/5

ro <r

^ UJ
3 •

® 2

P O

g-g

cz) a.

§ S

O a

;q o

5- f®

i’i
E ^

'o r.

^ S“

SS-

■ I

<T)

I ''

:x) DO 00 S s

^ » a. £. O

3 3:

3 g

ft)

pppppp-^7-

— LoLriLiboI— oJb^

vOOOJOOOC/iJi^'O
KJOOOO— ‘00-P.0^^0

" “ ^ - 1 g- 1 1=

^ g
1 K

— 3“ D-

p p p p ^ 7- 7- ;— to

k)*-P■b^bo^4s>■l/^L1^

ooa^uJo^c^oo^l>J

OO-OU)'0-OO-OUJ

p p p — ^ ^ OJ

*►— to ON to l/> ^ *—*

On ^O to vO O 01 to

-O to On to O 00 C/»

C/5

ft)

C/3

DO ft*

g* P

(/5 CL

g S

2. o

^ i

“. S'

= 3“

VOXJ

T* o

S 2

• CL

253

254

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

Figure 2. Quadratic regression showing a high degree of correlation for size classes of
woody plants with positive Ivlevs’ Indices of Selectivity most frequently foraged by
American beavers on central Texas rivers.

Discussion

Genera of woody plants foraged by beavers in central Texas
were similar to important foods in beaver diets in other regions of
North America (Svendsen 1980; Hill 1982; Roberts & Amer 1984;
Kindschy 1985; Beier & Barrett 1987; Baker & Hill 2003).
Although availability and species of these plant genera vary
geographically, they are consistently preferred as important food
resources for beavers. Deciduous woody species are often the most
important component of the diet of beavers; however, there is wide
regional variation in the number and composition of plants used
with as few as three species in the northern range but more species
in the southern range (Aleksiuk 1970; Baker & Hill 2003). The 18
different woody species used by beavers in this study were similar in
number to the 22 species reported in Louisiana but fewer than the 38

BACCUS, KAINER & SMALL

255

Table 4. Ivlev's Index of Electivity values among size classes of woody plants used by American
beavers on the San Marcos, Blanco and San Gabriel rivers in 1990-1991.

Species

Size Class (cm)

1 to 5

6 to 9

10 to 13

14 to 17

18 to 21

>21

Eastern cottonwood

0.147

-0.335

-1.000

-0.164

0.003

0.252

Black willow

0.052

-0.030

-0.120

-0.430

-0.200

0.003

Boxelder

0.094

0.004

-0.636

-0.953

-1.000

Hackberry

0.232

-0.114

-1.000

Pecan

0.261

0.225

0.173

-0.272

-0.126

-0.413

Sycamore

0.128

0.072

-1.000

-0.873

-0.782

-0.515

Dogwood

-0.005

0.028

-0.028

Privet

0.063

-0.044

-0.121

-0.538

-0.704

Mulberry

0.057

0.030

-0.277

-0.636

-0.845

-0.355

Texas ash

0.154

-0.131

-0.842

-0.787

-1.000

-0.858

Mexican buckeye

-0.017

0.085

0.061

Bald cypress

-0.573

-0.488

-1.000

-0.431

-0.035

0.204

American elm

0.088

-0.289

-0.242

-1.000

-0.062

0.115

Cedar elm

0.283

-0.407

-1.000

-0.153

-0.554

Bur oak

0.119

-0.335

-1.000

-0.069

Chinaberry

0.122

0.063

-1.000

Buttonbush

0.107

-0.225

Red Buckeye

0.051

-0.024

-1.000

species in South Carolina (Hill 1982). This study demonstrated two
basic patterns in beaver trophic ecology. Niche breadth for woody
plants did not vary among rivers, but differences in usage, relative
preferences, and size class use within those species did vary.

Niche is not defined solely by food resource selection, but diet is a
major concept in niche delineation (Krebs 1999). Eastern
cottonwood, white mulberry, Mexican buckeye, black willow,
boxelder, and bur oak had consistent usage in quantities >20% on all
rivers, however, frequency of use varied among rivers. This variation
is most likely explained by differences in species availability (Fryxell
& Doucet 1993). Although availability for eastern cottonwood was
minimal on the three rivers, the species had the highest preference by
beavers, greatest frequency of use on all rivers, and may represent the
southern counterpart of aspen in northern regions of North America.
Beavers inhabiting rivers of central Texas are opportunistic foragers,
consuming large quantities of other woody plants typically avoided
based on availability. Many frequently used woody plants in central
Texas were staple foods. Although these species were important diet

256

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

components, their usage was much lower than expected relative to
availability.

Beavers are known to select different food items at different
distances from their lodge (Jenkins 1980). Foraging distance (110 m)
for white mulberry was about twice the maximum distance (60 m) for
foraging in Alaska (Boyce 1981). The relative preferences for woody
species among rivers in this study were consistent with other
investigations of spatial variation in preferences for woody resources
by beavers (Henry & Bookout 1970; Jenkins 1981; Belovsky 1984;
Busher 1996; Barnes & Mallik 1997; Fryxell 2001). Although the
majority of woody plants occurred within 5 m of rivers, species
preference patterns varied spatially among rivers. Many woody plants
were consistently preferred while others were generally avoided
across all sites.

Although differences in spatial patterns of selection for woody
plants by beavers have been described (Henry & Bookout 1970;
Jenkins 1981; Belovsky 1984; Busher 1996; Barnes & Mallik 1997;
Fryxell 2001), little is known of the mechanism of selection when
availability of a species is variable. Beavers in central Texas have a
greater preference for sycamore, Texas ash, and rough-leaf dogwood
only when availability is low relative to other species. Sycamore
composed about 1% of total available woody stems on the San
Marcos River but was selected as a food with a frequency of 23.5%.
In contrast, sycamore was avoided on the Blanco River where it
comprised 17% of available woody plants. Beavers preferred Texas
ash (7% availability) on the Blanco River but avoided this species on
the San Gabriel River (28% availability). Similarly, rough-leaf
dogwood was avoided on the San Marcos River despite an availability
of 18%. Yet, on the Blanco River beavers showed a strong preference
for rough-leaf dogwood (availability < 1%). Different usages of the
same species on different rivers in this study may result from adaptive
diet selection in foraging patterns by beavers faced with different
availabilities of food resources on different rivers (Fryxell & Doucet
1993).

Patterns in relative preferences for size classes by beavers were
independent of river in central Texas. Beavers consistently selected

BACCUS, KAINER & SMALL

257

smaller stems in 15 of 18 woody species; however, beavers selected
stems of eastern cottonwood, black willow, bald cypress, and
American elm >21 cm in diameter. A similar trend of class-size usage
was recorded in Ohio (Nixon & Ely 1969), California (Hall 1960),
and Michigan (Belovsky 1984). Analysis of composition of limbs in a
beaver dam in New York showed the diameter of the largest limb was
about 16 cm (Shadle 1954), which was smaller than the largest tree
gnawed on central Texas rivers. However, no beaver dams were
observed in any of the study areas, and all gnawing activity was
assumed to be in the acquisitioning of food. Lower provisioning costs
may be the reason beavers generally selected smaller size classes of
woody plants (Jenkins 1980) because beavers require more energy to
transport larger stems to the central place (Fryxell & Doucet 1991;
1993). It is suspected that nutritional value of forage may be a factor
in forage selection in this study. However, relative preferences for the
5 to 9-cm size class of rough- leaf dogwood and the >21 -cm size class
of eastern cottonwood and bald cypress were primarily related to plant
availability and proximity to the river (Belovsky 1984).

Beaver colonies in central Texas occurred within or near large
quantities of preferred winter forage with a high diversity of
vegetative types but exhibiting low equitability of species. Beavers
select sites that have abundant winter foods and low numbers of other
species and areas with relatively high spatial heterogeneity of
vegetative types (Boyce 1981; Barnes & Mallik 1997). Similar
patterns of site selection occurred in central Texas. Plant community
structure of forage stretches shared one of two important
characteristics; relatively low equitability of woody species or diverse
vegetative types. Beavers also selected sites with an abundance of
small size (1-5 cm) woody plants. Further studies of beaver foraging
in central Texas are needed to compare woody plant characteristics of
foraged areas with surrounding areas via random sampling of forage
and non-forage stretches along rivers.

The ecological implications of selective foraging by beavers on
availability and structure of woody resources are significant because
beavers show distinct preferences for certain plants. With beaver
populations increasing throughout most of their range in response to

258

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

lowered trapping pressure, lack of natural predators, and harvest
restrictions (Novak 1987), land managers face potential problems
associated with herbivores. Effective beaver habitat management
requires an understanding of beaver foraging dynamics. This study
provides the first information on woody plant usage and selection by
beavers in central Texas. As a keystone species (Paine 1969) and an
ecosystem engineer (Jones et al. 1997), beavers influence species
composition and physical structure of riparian zones along central
Texas rivers.

Acknowledgments

D. G. Huffman assisted in the development of analytical methods.
D. H. Johnson at the United States Fish and Wildlife Service,
Northern Prairie Research Center provided valuable insights on
sampling design. D. Lemke provided expertise on plant identification.
R. Simpson. D. G. Huffman, and D. Lemke made helpful comments
on the manuscript. R. Welch, K. Dees, W. Collins, and R. Rolig
assisted with field work. We thank E. Cummings and C. C. Holt for
allowing access to their private property. Two anonymous reviewers
provided insightful comments which improved the original
manuscript. Texas State University and Texas Parks and Wildlife
Department provided funding.

Literature Cited

Aleksiuk, M. 1970. The seasonal food regime of arctic beavers. Ecology, 51:264-270.
Baker, B. W. & E. P. Hill. 2003. Beaver Castor canadensis. Pp. 288-310, in Wild
mammals of North America, biology, management, and economics (G. A.
Feldhamer, B. C. Thompson, and J. A. Chapman, eds.). The Johns Hopkins
University Press. Baltimore, Maryland. 1216 pp.

Barnes, D. M. & A. U. Mallik. 1997. Habitat factors influencing beaver dam
establishment in a northern Ontario watershed. J. Wildl. Manag., 61:1371-1377.
Basey, J. M., S. H. Jenkins & P. E. Busher. 1988. Optimal central-place foraging by
beavers: tree-size selection in relation to defense chemicals of quaking aspen.
Oecologia, 76:278-282.

Beier, P. & R. H. Barrett. 1987. Beaver habitat use and impact in the Truckee River
Basin, California. J. Wildl. Manag., 51:794-799.

Belovsky, G. E. 1984. Summer diet optimization by beaver. Amer. Midi. Nat., 1 1 1 :209-
222.

Blair, R. M. & M. J. Lauglinais. 1960. Nutria and swamp rabbit damage Taxodium
distichum seedlings. J. For., 58:388-389.

BACCUS, KAINER & SMALL

259

Boyce, M. C. 1981. Habitat ecology of an unexploited population of beavers in interior
Alaska. Pp. 155-186, in Proceedings, worldwide furbearer conference 1980 (J. A.
Chapman and D. Pursley, eds.). Worldwide Lurbearer Conference, Inc., Lrostburg.
Maryland. 2056 pp.

Busher, P. E. 1996. Pood caching behavior of beavers {Castor canadensis): selection
and use of woody species. Amer. Midi. Nat., 135:343-348.

Denney, R. N. 1952. A summary of North American beaver management. Colorado
Pish Game Department Report 28, Colorado Division of Wildlife. Denver, Colorado.
Donkor, N. T. & J. M. Pryxell. 1999. Impact of beaver foraging on structure of lowland
boreal forests of Algonquin Provincial Park, Ontario. Lor. Ecol. Manag., 1 18:83-92.
Donkor, N. T. & J. M. Pryxell. 2000. Lowland boreal forests characterization in
Algonquin Provincial Park relative to beaver {Castor canadensis) foraging and
edaphic factors. Plant Ecol., 148:1-12.

Doucet, C. M & J. M. Pryxell. 1993. The effect of nutritional quality on forage
preference by beavers. Oikos, 67:201-208.

Pryxell, J. M. 1992. Space use by beavers in relation to response abundance. Oikos,
64:474-478.

Pryxell, J. M. 2001. Habitat suitability and source-sink dynamics of beavers. J. Anim.
Ecol., 70:310-316.

Pryxell, J. M. & C. M. Doucet. 1991. Provisioning time and central place foraging in
beavers. Can. J. Zook, 69:1308-1313.

Pryxell, J. M. & C. M. Doucet. 1993. Diet choice and the functional response of
beavers. Ecology, 74:1297-1306.

Gallant, D., C. H. Berube, E. Tremblay & L. Vasseur. 2004. An extensive study of the
foraging ecology of beavers {Castor canadensis) in relation to habitat quality. Can.
J. ZooL, 82: 922-933.

Gould, F. W., G. O. Hoffman & C. A. Rechentin. 1960. Vegetational areas of Texas.

Texas A&M University and Agricultural Experiment Station, College Station, Texas.
Groeger, A. W., P. F. Brown, T. E. Tietjen & T. C. Kelsey. 1997. Water quality of the
San Marcos River. Texas J. Sci., 49:279-294.

Hall, E. R. 1981. The mammals of North America. John Wiley & Sons, New York,
New York. 2 Vols. 1181 pp.

Hall, J. G. 1960. Willow and aspen in the ecology of beaver on Sagehen Creek,
California. Ecology, 41:484-494.

Harper, J. L. 1969. The role of predation in vegetational diversity. Brookhaven Symp.
Biol., 22:48-62.

Henderson, F. R. 1960. Beaver in Kansas. Univ. Kansas Mus. Nat. Hist. Misc. Pub.
26:1-85.

Henry, D. B. & T. A. Bookout. 1970. Utilization of woody plants by beavers in
northeastern Ohio. Ohio J. Sci., 70:123-127.

Hill, E. P. 1982. Beaver {Castor canadensis). Pp. 256-281, in Wild mammals of North
America, biology, management, and economics (J. A. Chapman and G. A.
Feldhamer, eds.). The John Hopkins University Press. Baltimore, Maryland. 1216

pp.

Hurlbert, S. H. 1978. The measurement of niche overlap and some relatives. Ecology,
59:67-77.

Jenkins, S. H. 1975. Food selection by beavers: a multidimensional contingency table
analysis. Oecologia (Berlin), 21:157-173.

260

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

Jenkins, S. H. 1978. Food selection by beavers: sampling behavior. Breviora, 447:1-6.
Jenkins, S. H. 1979. Seasonal and year-to-year differences in food selection by beavers.
Oecologia, 44: 1 1 2- 1 1 6.

Jenkins, S. H. 1980. A size-distance relation in food selection by beavers. Ecology,
61:740-746.

Jenkins, S. H. 1981. Problems, progress, and prospects in studies of food selection by
beavers. Pp. 559-579, in Proceedings worldwide furbearer conference 1980 (J. A.
Chapman and D. Pursley, eds.). Worldwide Furbearer Conference, Inc., Frostburg.
Maryland. 3 Vols. 2,056 pp.

Jenkins, S. H. & P. E. Busher. 1979. Castor canadensis. Mamm. Species, 1-8.

Johnson, C. E. 1927. The beaver in the Adirondacks: its economics and natural history.
Roosevelt Wild Life Bull., 4:501-641.

Johnson, D. H. 1980. The comparison of usage and availability measurements for
evaluating resource preference. Ecology, 61:65-71.

Jones, C. G., J. H. Lawton & M. Shachak. 1997. Positive and negative effects of
organisms as physical ecosystem engineers. Ecology, 78:1946-1957.

Kindschy, R. R. 1985. Response of red willow to beaver use in southeastern Oregon. J,
Wildl. Manag., 49:26-28.

Krebs, C. J. 1999. Ecological methodology. Harper & Row, New York, New York.

620 pp.

McGinley, M. A. & T. G. Whitham. 1985. Central place foraging by beavers {Castor
canadensis)-, a test of foraging predictions and the impact of selective feeding on the
growth forms of cottonwoods {Popnlus fremontii). Oecologia, 66:558-562.

Nixon, C. M. & J. ELY. 1969. Foods eaten by a beaver colony in southeast Ohio. Ohio J.
Sci., 69:313-319.

Novak, M. 1987. Beaver. Pp. 283-313, w Wild furbearer management and conservation
in North America (M. Novak, J. A. Baker, M. E. Obbard, and B. Malloch, eds.).
Ashton-Potter Limited. Concord, Ontario. 1 150 pp.

Paine, R. T. 1969. A note on trophic complexity and community stability. Amer. Nat.,
103:91-93.

Pinkowski, B. 1983. Foraging behavior of beaver {Castor canadensis) in North Dakota.
J. Mammal., 64:312-315.

Roberts, T. H. & D. H. Amer. 1984. Food habits of beaver in east-central Mississppi. J.
Wildl. Manag., 48:1415-1419.

Schmidly, D. J. 2004. The mammals of Texas. The University of Texas Press, Austin,
Texas. 501 pp.

Shadle, A. R. 1954. Sizes of wood cuttings handled by beavers. Amer. Midi. Nat.,
52:510-512.

Slough, B. G. & R. M. Sadlier. 1977. A land capability classification system for beaver
{Castor canadensis). Can. J. Zook, 55:1324-1335.

Svendsen, G. E. 1980. Seasonal change in feeding patterns of beaver in southeastern
Ohio. J. Wildl. Manag., 44:285-290.

Tupa, D. D. & W. K. Davis. 1976. Population dynamics of the San Marcos salamander,
Eurycea nana Bishop. Texas J. Sci., 27:179-195.

Zar, J. H. 1984. Biostatistical Analysis (2ond ed.). Prentice-Hall, Inc., Englewood
Cliffs, New Jersey. 720 pp.

JTB at: jb02@txstate.edu

TEXAS J. OF SCL 59(4):26 1-276

NOVEMBER, 2007

EFFECTS OF PERMANENT WATER
ON HOME RANGES AND MOVEMENTS OF
ADULT MALE WHITE-TAILED DEER IN SOUTHERN TEXAS

Stephen L, Webb*, David G. Hewitt and Mickey W. Hellickson**

Caesar Kleberg Wildlife Research Institute, MSC 218
Texas A&M University-Kingsville, Kingsville, Texas 78363 and
**King Ranch, Inc., P.O. Box 1090, Kingsville, Texas 78364
^Present address:

Department of Wildlife and Fisheries
Mississippi State University, Box 9690
Mississippi State, Mississippi 39762-9690

Abstract. -Water distribution can have a significant impact on ungulates in arid
environments if water is scarce. When water distribution is inadequate, then
ungulates may have large home range sizes, or be concentrated near water sources.
Because establishing water sources is expensive, research is needed to determine
distribution of water sources adequate to meet management goals. To assess the
effect of water distribution on white-tailed deer (Odocoileus virginianus) in a semi-
arid environment, 48 adult (estimated > four years of age), male white-tailed deer
were captured, radio-collared and their movements tracked over two years in southern
Texas. Densities of ponds and concrete troughs in deer home ranges were 5 to 34%
and 1 7 to 62% lower than what was available on the study area, respectively. There
were more ponds (0.409 water sources/km^) than concrete troughs (0.125 water
sources/km^) in deer home ranges and deer were found 397-442 m closer to ponds
compared to concrete troughs. Most often deer were found 500-1,000 m from the
nearest water source during all seasons. There was no difference (P > 0.1047) in
distance between actual telemetry locations and random locations to the nearest water
source. These data show that water distribution was adequate during the two years of
the study. However, during drier years, the distribution and density of water sources
might be more critical to white-tailed deer movements.

Water distribution may have a large impact on ungulates in arid
environments if water is scarce. If water distribution is poor, then
ungulates may have large home range sizes, or be concentrated near
water sources. Because establishing water sources can be
expensive (Bone et al. 1992; Broyles 1998; Gunn 1988; Mouton &
Lee 1992), it is important to determine the minimum number of
watering sources necessary to support animals in an area, so that
resources are not wasted establishing water sources.

262

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

There are many proponents and critics of wildlife water
developments. Proponents of water developments site scarcity of
free-standing water as a limiting factor of wildlife populations in
arid environments (Roberts 1977). Critics, on the other hand,
suggest that addition of new water sources may not return expected
benefits (Broyles 1995; Brown 1998). If new water supplies are to
yield expected benefits then water must be limiting to begin with,
new sources must be distributed adequately in usable habitat, and
they must be sources that wildlife are willing to use. The placement
of concrete troughs near roads and human disturbances and the
influence of cattle may reduce their usefulness to deer and other
wildlife species. Prasad & Guthery (1986) found that deer did not
use concrete troughs due to higher human and cattle activity. Ponds
are often built on ephemeral drainages, which often contain deeper
soils and an abundance of vegetation, resulting in higher levels of
screening cover. Therefore, deer may water at ponds instead of
concrete troughs due to less human influence and their larger size.

In southern Texas, water may be limited for portions of the year
or for several consecutive years because of frequent and widespread
droughts. The resulting lack of free-standing water may be
deleterious to white-tailed deer (Odocoileus virginianus) because
intake of water is essential for body condition maintenance
(Maghini & Smith 1990), deer eat less and lose body weight when
deprived of water, and are not adapted to dehydration (Lautier et al.
1988).

The location and abundance of water has been found to affect
distribution and behavior of white-tailed deer. In Arizona, when
water became scarce in June, white-tailed deer moved closer to
permanent water, but dispersed when summer rains started (Welch
1960). Maghini & Smith (1990) found that deer would move from
established diurnal ranges to distant water sources. Visual
observations of deer in the Coastal Prairies of Texas found deer
movements may have been attributed to the location of bodies of
water (Michael 1965). Because Michael’s (1965) data were based

WEBB, HEWITT & HELLICKSON

263

on visual observations, it appeared deer used areas in association
with water, yet water sources were within 686 m of one another.
Therefore, deer were never far from water at any time.

Deer are thought to need free-standing water; therefore, water
availability in xeric environments probably influences home range
size and location (Maghini & Smith 1990). The importance of
drinking water in the life of deer was emphasized by the fact that
watering sites were frequently centers of deer home ranges and the
presence or absence of water noticeably affected their daily
activities (Michael 1968). White-tailed deer are sedentary and
reluctant to leave an area; however, during dry seasons when
surface water becomes scarce, deer may concentrate near remaining
sources of water (Michael 1968). Teer (1996) reported white-tailed
deer used water sources 2-3 times/day. Therefore, water appears to
exert a strong Influence on daily activity patterns, movements, and
home ranges.

In contrast, when water distribution was adequate to meet needs,
mule deer {Odocoileus hemionus) distribution probably was
determined by environmental factors other than water (Boroski &
Mossman 1996). Webb et al. (2006) found that white-tailed deer
visits to permanent water following rainfall events decreased for
three weeks indicating that deer obtained water from ephemeral
pools or from vegetation. Camera surveys conducted in
conjunction with Webb et al. (2006) found no ear-tagged (i.e., of
>150 marked) or radio-collared deer {n = 48) using water during
two summers of surveys. Thus, in southern Texas, permanent
sources of water may not exert as strong an influence on deer as in
other arid environments.

If deer need to expand or shift home ranges when water is
limiting, then these expansions or shifts will help dictate the scale at
which deer management needs to occur. The objectives of this
study were to: (1) determine the effect of permanent water sources
on annual home range formation of adult, male deer, and (2)

264

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

determine how seasonal and annual deer movements were affected
by permanent sources of water.

Study Area

This study was conducted on the Callaghan Ranch (27°48'59"N,
99°18’49'’W), from September 2002 through August 2004, in Webb
County, Texas, 43.4 km northeast of Laredo, Texas. The ranch had
no deer-proof fences and consisted of 34,400 ha of mesquite-
{Prosopis glandulosa) dominated shrubland (McCoy et al. 2005).
Principle soils of the Callaghan Ranch were Catarina-Maverick-
Moglia which are deep to moderately deep, nearly level to gently
rolling with saline, clayey, and loamy soils (Sanders & Gabriel
1985). The ranch was continuously stocked with domestic cattle at
a rate of 1 animal unit/21 ha and deer density was estimated to be 1
deer/11.4 ha in 2002 and 1 deer/12.4 ha in 2003. Free water was
distributed across the ranch through earthen water catchments,
troughs, and ephemeral creeks at an average density of 0.83
permanent water sources/km^ (Webb 2005). Feeding stations with
pelleted, supplemental feed occur on the study area at an average
density of 1 feeder station/121.5 ha. Com was provided as bait at
hunters’ discretion during the hunting season from November-
January.

Webb County is in the western Rio Grande Plains of Texas with
elevations from 122-274 m above sea level. Cattle ranching and
leasing hunting rights for white-tailed deer and other game are
primary economic activities. Average relative humidity is 60% at
midaftemoon, 80% at dawn, and usually higher at night. Webb
County had an average daily maximum temperature in July of
36.8°C, an average daily minimum in January of 6.3°C, and
receives a mean annual rainfall of 50.3 cm (Sanders & Gabriel
1985). Most rainfall (70%) usually falls between April and
September (Sanders & Gabriel 1985). Total annual rainfall was
48.7, 65.2 and 69.1 cm in 2002, 2003 and 2004, respectively
(Laredo, Texas; National Climatic Data Center, 2003-2005).

WEBB, HEWITT & HELLICKSON

265

Methods

Capture and -Nineteen adult (estimated > four years of

age) male, white-tailed deer were captured on the Callaghan Ranch
during October 2002 using a net-gun fired from a helicopter
(Barrett et al. 1982; Webb et al. 2008). An additional 13 males
originally captured and radio-collared as yearlings in 1998-1999 as
part of a study on white-tailed deer dispersal (McCoy et al. 2005)
were also captured. Three additional males were captured in
October 2003 and 13 additional males in March 2004. To minimize
mortalities due to capture myopathy, the helicopter did not pursue
deer more than eight minutes (DeYoung 1988).

Deer were blindfolded once captured, manually restrained, and
aged according to tooth replacement and wear (Severinghaus 1949).
The 13 deer originally radio-collared as yearlings were of known
age and provided site-specific tooth wear patterns that helped
ensure subsequent captures represented adult males. Deer were ear-
tagged with two color-coded, numbered tags representing the year
of capture and the deer’s estimated age. Deer were fitted with
radio-collars equipped with both activity and mortality sensors
(Advanced Telemetry Systems, Inc., Isanti, Minnesota) and
released at site of capture within 20 min to minimize stress.
Capture and handling procedures were approved by the Texas
A&M University-Kingsville Institutional Animal Care and Use
Committee (Permit No. 2003-5-14).

Radio-telemetry. -Dqqv were located one to two times/week
using a Telonics TR-2 radio-receiver with TS-1 scanner (Telonics,
Inc., Mesa, Arizona) and a null-peak radio telemetry system
consisting of two yagi 4-element antennas (Advanced Telemetry
Systems, Inc., Isanti, Minnesota). Bearings to the transmitter were
estimated using the null-peak radio telemetry system and a hand¬
held compass. Compass bearings were taken >12 m from the truck
to reduce interference and then corrected for declination (White &
Garrott 1990) by adding 6.5° to the final bearing.

266

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

Diurnal, nocturnal, and crepuscular locations were estimated
using two to five bearings and Location of a Signal software
(LOAS; Ecological Software Solutions™, Sacramento, California).
Locations derived from > three bearings were estimated by the
Maximum Likelihood Estimator (Lenth 1981) and those derived
from two bearings were calculated by Best Biangulation Estimator
in LOAS. Locations were converted for use in a Geographic
Information System (GIS) and Arc-View 3.2 software
(Environmental Systems Research ' Institute, Inc., Redlands,
California). Relocations on deer were obtained >12 hours apart to
allow deer sufficient time to travel across its home range. Visual
observations were recorded with a differential GPS (DGPS) unit
and were used in the final estimation of home range size.

To assess accuracy of the null-peak radio telemetry system,
radio transmitters were randomly placed throughout the study area
and georeferenced with a DGPS unit. Mean telemetry error (N =
13) was 81.8 ± 1 1.7 m for locations with > three bearings.

Data collection and analysis -T\\q fixed kernel home range
estimator (Worton 1989) was used to generate 95% home ranges
using the Animal Movement Extension (Hooge & Eichenlaub
1997) in ArcView 3.2 software. The reference bandwidth (href) was
used when calculating the 95% probability polygons. Annual home
ranges were calculated from 15 October 2002 to 14 October 2003
(year one) and 15 October 2003 to 25 August 2004 (year two).

Water sources were delineated within the study area using 1995
Digital Ortho Quadrangle maps, or mapped using a DGPS and
brought into ArcGIS 8.2 software. Water sources were classified as
concrete troughs or earthen ponds and monitored to determine if,
and when, they went dry. Only two of the monitored water sources
went dry for two weeks during summer 2003. Therefore, water
sources holding water for >11.5 months/year were considered
permanent. Neither of the water sources that went dry were within
the home ranges of radio-collared deer.

WEBB, HEWITT & HELLICKSON

267

The study area was defined by creating a Minimum Convex
Polygon around all deer telemetry locations and buffering out 82 m
(average error of the null-peak telemetry system in this study). One
deer was excluded from the analysis because it was found on a
neighboring property, thus it was not possible to map all water
sources. If the study area fell outside of ranch boundaries then the
study area was redrawn to conform to the ranch boundary.

The distance (m) from each telemetry location to the nearest
water source of each type, within a season (i.e., spring, summer,
fall, winter), was calculated for each deer using the spatial join
function in ArcGIS 8.2 software. This analysis did not involve deer
home ranges, only telemetry locations within season. Deer with >
five locations/season were used in this analysis, even if deer were
not tracked for one full year. Therefore, all telemetry locations
within a season were used to determine mean distance to each
respective water source so the effects of the two types of water
sources on deer movements could be measured. All water types
were pooled and a mean distance to the nearest water source for
each deer during each season was calculated.

The random point generator in ArcGIS 9 was used to generate
1000 random points to determine whether deer were found closer
than expected to each type of water source within each deer’s home
range. Only deer that were tracked one full year and that had >30
locations were used in the analysis when generating random
locations within home ranges. All random locations within a home
range were used to determine mean minimum distance to each type
of water source. Mean minimum distance of random points was
compared to the mean minimum distance of actual telemetry
locations for each season to determine if actual deer locations were
closer to water sources within the home range.

Statistical analysis -A 1 -sample Mest was used to assess how
water sources affected the establishment of home ranges within the
study by comparing the densities (water source/km^) of water

268

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

sources within home ranges to the density available within the
study area. Because the density of water sources within the study
area changed, through the construction of additional water sources
at the end of year one, the two years were analyzed separately. A
2-sample /-test was used to assess home range size differences for
deer with water and without water in their home range.

A randomized complete block design (RCBD) analysis of
variance (ANOVA), with deer as blocks, was used to test for
seasonal differences within year for distance to nearest pond,
concrete trough, and water source. A RCBD ANOVA was also used
to test if actual seasonal distances to water sources differed from
random location distances for each respective type of water source.
Distance between ponds and concrete troughs were compared using
a paired /-test after data was pooled across seasons within year.
Distances to ponds and distances to concrete troughs were also
pooled within home ranges across seasons within a year and
compared to random locations using a paired /-test. Last, the
percent of locations was calculated for each deer, regardless of
whether they had a water source within their home range, to the
nearest water source within 5 distance categories (0-500, 500-1,000,
1,000-1,500, 1,500-2,000, and >2,000 m). All analyses were
conducted using SAS 9.1 (SAS Institute, Cary, North Carolina).
Statistical significance was concluded for P <0.05. Means are
reported ± SE.

Results

During this study, 48 adult, male white-tailed deer were
monitored. Not all deer survived an entire year; therefore, they
were excluded from annual home range analyses. This left 38 deer
(26 deer in year one and 12 deer in year two) in the study for a total
of 16,325 radio-days (10,028 radio-days in year one and 6,297
radio-days in year two). Deer ranged in age from 4.5 to >8.5 years-
of-age.

WEBB, HEWITT & HELLICKSON

269

Average number of relocations used during years one and two
were 56 ± 1.3 and 40 ± 1.2, respectively. Average 95% fixed
kernel home range size was 207.4 ± 20.4 ha and 225.7 ± 30.1 ha for
years one and two, respectively (Webb et al. 2007). Deer with
water in their home range had larger home range sizes (227.7 ±
21.4 ha) than deer without (185.1 ± 25.5), but was not statistically
different (df= 36, / = -1.22, P = 0.231).

The average density of ponds within deer home ranges was five
and 34% lower in years one and two, respectively, than the average
density of ponds on the study area for both years (Table 1). The
average density of concrete troughs within deer home ranges was
62 and 1 7% of the density of concrete troughs on the study area in
years one and two, respectively (Table 1). Density of ponds within
home ranges was greater than the density of concrete troughs in
home ranges during both years. The density of concrete troughs
within home ranges for both years (0.196 ± 0.056 and 0.054 ±
0.054 water sources/km^) was significantly (P < 0.047) lower than
the density of troughs across the study area (0.314 water
sources/km^). There was no difference (P > 0.179) in densities of
ponds within home ranges (0.470 ± 0.089 and 0.349 ± 0.126 water
sources/km^) compared to the study area (0.496 and 0.529 water
sources/km^) during either year.

During year one, 27% (7 of 26) of deer did not have any
permanent water source within their home range. Fifty percent (6
of 12) of deer in year two did not have any permanent water source
within their home range. However, all of these deer had a
permanent water source within 499 m of their home range
boundary.

When all water sources were combined for both years, 26.4,
48.2, 15.0, 7.9, and 2.6% of locations were in distance groups 0-
500, 500-1,000, 1,000-1,500, 1,500-2,000, and >2,000 m,

respectively. Most deer were found within 500-1,000 m of all types
of water sources compared to any other distance for all seasons and

Table 1. Mean density of water sources (water sources/km^) within deer home ranges and within study area from 15 October 2002-14
October 2003 (Year 1) and 15 October 2003-25 August 2004 (Year 2) on the Callaghan Ranch in southern Texas. Twenty-six deer
were used in the analysis in year 1 and 12 deer in year 2.

Study Area Home Range _

270

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

0\ r-- On — ^
VO r- o
o o
o o o o

O ON ^

m o

O (N Tf

ON NO NO
00 fN NT)
0 0-^0
O O O O

O NO ON
ON vn

' m o
o' o o o

VO On

On <N —

Tf r<N vn

o o o o

— ^ — (N <N

(/)

(50
W o
•O o
c ^
o H

Ci. (U

^ o

i §1 §

UJ U W U

S 3
Q

^ •

3 «} >,

2=3

« C)

o ' "

c
o
o

c/T

*0
3

O „
<=^ o
B

(U c

I =

SI
X s

£ w

(50

.£

^ §
<u m
(50 O

c o

S rM

-o
o
o

i 8

-r-l

£ 1

3 (U

£ c

.£ <u

c (5r

w O

o x>

S §

(N ’O ^

•o

c<3

Cei

IT)

.—1

r--

1— 1

f--

.—1

-H

■M

CNJ

'If

■'d-

T— 1

-H

'sf

-H

r--

I— t

-H

C-'

'If

>— 1

-H

-F

-H

—

I— 1

.—1

— -G — -C

ns 00 T3

3 3 — 3 3 ~

O O < p O <

fi: fi:

— ^ CN <N <N

Comparison of seasonal minimum distance means for each respective water type was tested using

Comparison of seasonal and random minimum distance means for each respective water type was tested using y4A^OF/4.

WEBB, HEWITT & HELLICKSON

271

years. Averaged across years, 75% of deer locations were within
1,000 m of a permanent water source, and >89% were within 1,500
m of a permanent water source. The farthest deer were located
from water at any time was 2,872 m, and the closest was 0 m.

In both years, there was no difference among seasons in the
distance between deer locations and ponds (P > 0.233), concrete
troughs (P > 0.764), or all water sources combined (P > 0.415;
Table 2). Furthermore, there was no difference (P > 0.226)
between the distances deer were found from water sources
compared to the distance random points were from water sources
for any type of water source within any season (Table 2).

In both years, when data were pooled across seasons, deer were
closer (year one F\^23 = 10.03 P = 0.004; year two Fi^n = 9.63 P =
0.01) to ponds (year one - 761 ± 64 m; year two - 672 ± 84 m)
compared to concrete troughs (year one - 1,158 ± 104.0 m; year
two - 1,114 ± 120 m). In both years, there was no difference in
mean minimum distance between randomly generated locations
within home ranges and actual telemetry locations for ponds (P >
0.426) and concrete troughs (P > 0.105; Table 3).

Discussion

These data showed that home ranges occurred in areas with a
lower density of ponds and concrete troughs than on the study area
as a whole. The placement of concrete troughs near roads, around
areas of human disturbance, and the influence of cattle in these
areas may reduce their usefulness to deer and other wildlife species,
which agrees with the findings of Prasad & Guthery (1986). Even
though the study area contained higher densities of ponds compared
to deer home ranges, ponds were found in higher densities in deer
home ranges compared to concrete troughs. Ponds were built on
ephemeral drainages to catch runoff after rain events. The drainages
associated with ponds often contained deeper soils and an abundance
of vegetation. The abundance of vegetation resulted in higher levels

272

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

Table 3. Mean minimum distances (m) to earthen ponds and concrete troughs from
randomly generated locations within home ranges and actual telemetry locations
pooled across seasons within year on the Callaghan Ranch, Webb County, Texas for
2 years. During year 1 deer were tracked from 15 October 2002-14 October 2003 and
during year 2 from 15 October 2003-25 August 2004.

Year

Source

Random

Mean ± SE

Actual

Mean ± SE

r-value

P-value^

Pond

766 ±61

767 ± 66

-0.5

0.635

Trough

1,133 ± 100

1,156± 103

1.7

0.105

Pond

682 ± 86

736± 131

-0.8

0.426

Trough

1,107± 119

1,146± 118

0.7

0.489

^Comparisons of annual actual telemetry locations and randomly generated locations
were tested using a paired Atest.

of screening cover compared to areas around concrete troughs, which
may lead to increased use of ponds.

Michael (1968) postulated that watering sites are frequently in the
centers of home ranges and that water affects daily movements.
Under xeric conditions, water availability probably influenced white¬
tailed deer home ranges (Maghini & Smith 1990). However, 27-50%
of deer did not have a permanent water source within their home
range. If watering sites were not found within home ranges, then the
watering sites were on the periphery of the home range and it is likely
deer used these at times other than when deer were located. Home
range size of deer with water in their home range was larger than
those of deer without water in their home range. This may indicate
that deer with larger home range sizes had to expand their home range
to include water. The relatively small home range sizes of deer in this
study may explain why up to 50% of deer did not have water in their
home range.

Boroski & Mossman (1996) found when summer water
distribution was adequate to meet mule deer needs, the distribution of
deer was probably determined by other factors in the environment.
Using the density of water sources within home ranges compared to
the study area as an indicator of how deer were spaced throughout the
landscape revealed that during the two years of the study, water

WEBB, HEWITT & HELLICKSON

273

availability on the study area did not affect deer distribution within the
landscape. Maghini & Smith (1990) reported that deer that did not
have a water source in their home range did not expand their home
range to include one, but instead shifted their activity within their
home range closer to a water source. Therefore, to detect differences
in activity, actual telemetry locations within home ranges were used to
reveal how deer changed their movement patterns throughout the year.
Contradictory to Michael (1965; 1968) and Maghini & Smith (1990),
deer did not change their activity within their home range relative to
permanent water during any season, albeit the relocation frequency
within season was <20 relocations.

Distances deer were located from water generally reflected the
distribution of water sources (Boroski & Mossman 1996; Mackie
1970). Because the density of water sources on the study area was
relatively high (0.83 water sources/km^) deer were never far from a
water source. The distribution of water sources on the study area is
reflected in the average distance deer were located from them. This
may have a positive effect on deer because they do not have to make
long-distance movements to water which could conserve energy,
particularly during the summer, and reduce mortality. In the present
study, no difference was found between distance of deer locations to
water sources and random locations.

Water was readily available as a number of sources (i.e., ponds,
concrete troughs, ephemeral pools, creeks) during the present study.
So, when water distribution is adequate to meet deer needs, deer
distribution is probably influenced by other factors in the environment
(Boroski & Mossman 1996) such as habitat, feed, cattle, and other
wildlife species. The data from this study provides a measurement of
adequate water distribution for deer in southern Texas during
relatively wet years of above average annual rainfall. During dry
years, deer may require a greater density of water sources to meet
their needs. Water turnover rates are much higher in pregnant and
lactating animals (Macfarlane & Howard 1972; Maloiy et al. 1979)
because water requirements of lactating females are quite high (Short

274

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

1981). Therefore, females may require a higher density of water
sources spaced at closer intervals.

A basic question that arises when designing wildlife water systems
is, “how far apart should water sources be spaced?” This will dictate
how many water sources are needed on an area. It has been suggested
that a density of one water source/km^ should be more than adequate
(Maghini & Smith 1990). Demarais et al. (2000) suggested that one
water source/1 50-200 ha is sufficient. Data from this study was
interpreted to show a density of one reliable and usable water source/4
km^ should be sufficient to minimize movements to water based on
home range size and movements of white-tailed deer. Water sources
should be spaced 2 km apart to meet the density requirements (i.e.,
one water source/4 km^). Therefore, deer will never be farther than
1 .4 km from the nearest water source.

Managers and agencies in charge of water development projects
should provide reliable and usable water sources at the proper density
and spacing to minimize expenses and resources. This study provides
information on water distribution of two types for adult male white¬
tailed deer. Due to potential disturbance by cattle and humans at
concrete troughs, ponds for wildlife may provide better use of the
resource. Further studies should investigate actual dependence on
water of free-ranging white-tailed deer in arid environments during
wet and dry years and requirements for other ages and sexes.

Acknowledgments

We thank C. Rush and D. Lee for providing housing, J. B. Finley
and the Callaghan Ranch for access, F. Robbins and J. E. McCoy for
field assistance, D. Marquedt, D. Jones, and R. S. Lyons for volunteer
help and observational data, B. M. Ballard for use of equipment, and
R. L. Bingham for statistical analysis. We also thank T. E. Fulbright,
A. Ortega-Santos, K. L. Gee, and multiple anonymous reviewers for
helpful comments on earlier drafts. Cooperative funding was
provided by the San Antonio Livestock Exposition, Inc. and Houston
Safari Club. This is contribution number 06-128 of the Caesar
Kleberg Wildlife Research Institute.

WEBB, HEWITT & HELLICKSON

275

Literature Cited

Barrett, M. W., J. W. Nolan & L. D. Roy. 1982, Evaluation of a hand-held net-gun to
capture large mammals. Wildl. Soc. Bull., 10:108-114.

Bone, T, L., R. Cantu & S. Brownlee. 1992. Water for west Texas Wildlife. Eederal
Aid Project W-129-M. Texas Parks and Wildlife Department, Austin, Texas.

Boroski, B. B. & A. S. Mossman. 1996. Distribution of mule deer in relation to water
sources in northern California. J. Wildl. Manage., 60:770-776.

Brown, D. E. 1998. Water for wildlife: belief before science. Pp. 9-16 in Proceedings of
a symposium on environmental, economic, and legal issues related to rangeland
water developments. The Center for the Study of Law, Science and Technology,
Arizona State University, Tempe, Arizona, 622 pp.

Broyles, B. 1995. Desert wildlife water developments: Questioning use in the
Southwest. Wildl. Soc. Bull., 23:663-675.

Broyles, B. 1998. Reckoning real costs and secondary benefits of artificial game waters
in southwestern Arizona. Pp. 236-253 in Proceedings of a symposium on
environmental, economic, and legal issues related to rangeland water developments.
The Center for the Study of Law, Science and Technology, Arizona State University,
Tempe, Arizona, 622 pp.

Demarais, S., K. V. Miller & H. A. Jacobson. 2000. White-tailed deer. Pp. 601-628 in
S. Demarais and P. R. Krausman, editors. Ecology and Management of Large
Mammals in North America. Prentice-Hall, Inc., Upper Saddle River, New Jersey,
778 pp.

DeYoung, C. A. 1988. Comparison of net-gun and drive-net capture for white-tailed
deer. Wildl. Soc. Bull., 16:318-320.

Gunn, J. 1988. Water development planning data handout. Arizona Game and Fish
Department, Phoenix, Arizona.

Hooge, P. N. & B. Eichenlaub. 1997. Animal movement extension to Arc-View. Ver
1.1. Alaska Biological Center, U.S. Geological Survey, Anchorage, Alaska.
http://www.absc.usgs.gov/glba/gistools/animal_mvmt.htm

Lautier, J. K., T. V. Dailey & R. D. Brown. 1988. Effect of water restriction on feed
intake of white-tailed deer. J. Wildl. Manage., 52:602-606.

Lenth, R. V. 1981. On finding the source of a signal. Technometrics, 23:149-154.

Macfarlane, W. V. & B. Howard. 1972. Comparative water and energy economy of wild
and domestic mammals. Symposia of the Zool. Soc. of London, 31:261-296.

Mackie, R. J. 1970. Range ecology and relations of mule deer, elk, and cattle in the
Missouri River Breaks, Montana. Wildl. Monogr., 20:1-79.

Maghini, M. T. & N. S. Smith. 1990. Water use and diurnal seasonal ranges of Coues
white-tailed deer. Pp. 21-34 in P. R. Krausman and N. S. Smith, editors. Managing
Wildlife in the Southwest. Arizona Chapter of The Wildlife Society, Phoenix,
Arizona, 262 pp.

Maloiy, G. M. O., W. V. Macfarlane & A. Shkolnik. 1979. Mammalian herbivores. Pp.
185-209 in G. M. O. Maloiy, editor. Comparative physiology and osmoregulation in
animals. Volume 2. Academic Press, London, England, 246 pp.

McCoy, J. E., D. G. Hewitt & F. C. Bryant. 2005. Dispersal by yearling male white¬
tailed deer and implications for management. J. Wildl. Manage., 69:366-376.

Michael, E. D. 1965. Movements of white-tailed deer on the Welder Wildlife Refuge. J.
Wildl. Manage., 29:44-52.

276

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

Michael, E. D. 1968. Drinking habits of white-tailed deer in south Texas. SE
Association of Fish and Wildl. Agencies, 21:51-57.

Mouton, R. J. & R. M. Lee. 1992. Actual costs of bighorn sheep water developments.
Desert Bighorn Council Transactions, 36:49-50.

National Climatic Data Center. 2003-2005. Climatological data annual summary: Texas.
Volume 108-110. National Oceanic and Atmospheric Administration, United States
Department of Commerce, Asheville, North Carolina.
http://www.ncdc.noaa.gov/oa/ncdc.html

Prasad, N. L. N. S. & F. S. Guthery. 1986. Wildlife use of livestock water under short
duration and continuous grazing. Wildl. Soc. Bull., 14:450-454.

Roberts, R. F. 1977. Big game guzzlers. Rangeman’s Journal, 4:80-82.

Rosenstock, S. S., W. B. Ballard & J. C. Devos, Jr. 1999. Viewpoint: Benefits and
impacts of wildlife water developments. J. Range Manage., 52:302-31 1.

Sanders, R. R. & W. J. Gabriel. 1985. Soil survey of Webb County, Texas. Soil
Conservation Serviee, USDA, Washington D.C., 124 pp.

Severinghaus, C. W. 1949. Tooth development and wear as criteria of age in white¬
tailed deer. J. Wildl. Manage., 13:195-216.

Short, H. L. 1981. Nutrition and metabolism. Pp. 99-127 in O. C. Wallmo, editor. Mule
and black-tailed deer of North America. University of Nebraska Press, Lincoln,
Nebraska, 629 pp.

Teer, J. G. 1996. The white-tailed deer: natural history and management. Pp. 193-210
in P. R. Krausman, editor. Rangeland wildlife. The Society for Range Management,
Denver, Colorado, 440 pp.

Webb, S. L. 2005. Home range, mortality, and water use of mature, male white-tailed
deer in south Texas. Unpublished M.S. thesis, Texas A&M University-Kingsville,
Kingsville, Texas, 137 pp.

Webb, S. L., C. J. Zabransky, R. S. Lyons & D. G. Hewitt. 2006. Water use and quality
of wildlife watering sources during summer in Texas. SW Nat., 51:368-375.

Webb, S. L., D. G. Hewitt & M. W. Hellickson. 2007. Scale of management for mature
male white-tailed deer as influenced by home range and movements. J. Wildl.
Manage., 71:1507-1512.

Webb, S. L., J. S. Lewis, D. G. Hewitt, M. W. Hellickson & F. C. Bryant. 2008.
Assessing the helicopter and net-gun as a capture technique for white-tailed deer. J.
Wildl. Manage., 72:310-314.

Welch, J. M. 1960. A study of seasonal movements of white-tailed deer in the Cave
Creek Basin of the Chiricahua Mountains. Unpublished M.S. thesis. University of
Arizona, Tucson, Arizona, 79 pp.

White, G. C. & R. A. Garrott. 1990. Analysis of wildlife radio-tracking data. Academic
Press, San Diego, California, 383 pp.

Worton, B. J. 1989. Kernel methods for estimating the utilization distribution in home-
range studies. Ecology, 70:164-168.

SLW at: stephenwebb@hotmail.com

TEXAS J. OF SCI. 59(4):277-290

NOVEMBER, 2007

SPATIAL AND TEMPORAL EXPRESSION OF SYNDECAN - 2
(FIBROGLYCAN) DJ CHICK HEART DEVELOPMENT

Ashley Gordon and J. Kevin Langford

Department of Biology, Stephen F. Austin State University
Box 13006 SFA Station, Nacogdoches, Texas 75962

Abstract.-Heparan sulfate proteoglycans, including the syndecans, play impor¬
tant roles during development and wound healing. While syndecan-2 is expressed by
cells of the developing heart, as shown by Northern blot analysis, the developmental
pattern and cellular specificity of syndecan-2 expression remains unknown. Using
immunohistochemical analysis, syndecan-2 was found to be expressed within the
myocardium beginning at the onset of epicardial formation in the chick heart and
continued to be expressed in this tissue-specific manner throughout the stages
examined. With growth factors such as FGF, TGFp and BMP expressed in cardiac
tissue, syndecan-2 likely is playing a role in mediating the effects of these and other
factors which are critical for normal cardiac morphogenesis.

Syndecan-2 is but one member of the syndecan family of
heparan sulfate proteoglycans. Four syndecan isoforms occur in
higher vertebrates and a single isoform in invertebrates including
Drosophila and the nematode Caenorhabditis elegans. Regardless
of species or isoform, all syndecans share structural homology. In
addition to a hydrophobic transmembrane domain, the cytoplasmic
domain is organized into a membrane proximal constant domain
(Cl), a variable domain (V) with shared homology only between
species, and a COOH terminal constant domain (C2) ending with a
PDZ-binding motif (Lopes et al. 2006). The extracellular domain
(aka. ectodomain) is the most variable of the regions and is thought
to contribute to the distinct functions of the different syndecan
isoforms (Fears & Woods 2006). Recent evidence indicates that the
ectodomain core protein sequences are not only critical for the
specification and display of glycosaminoglycans (Zhang et al.
1995), but are also essential for mediating cellular activity (McFall
& Rapraeger 1998; Langford et al. 2005).

Syndecans are present on a wide variety of cell types and
commonly exhibit a developmentally regulated pattern (Rapraeger
2001). Functional studies demonstrate that syndecans mediate

278

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

cellular activities including cell adhesion, binding and modulation
of growth factor activity, interacting with and arranging the
cytoskeleton and organizing of the extracellular matrix. These
processes are critical during morphogenesis and become reiterated
in tissues undergoing wound healing following injury.

During cardiac morphogenesis, many of the biological activities
mediated by syndecans are active in shaping cardiac tissue into a
functioning complex organ. Cellular proliferation and differentia¬
tion are likely regulated to a large degree by the array of growth
factors present, including basic fibroblast growth factor (bFGF)
(Sheikh et al. 2004), transforming growth factor beta (TGFp)
(Akhurst et al. 1990) bone morphogenetic protein (BMP) (Neuhaus
etal. 1999) and thrombospondin (Corless et al. 1992). Construction
of the extracellular environment and directed cell migration are
additional processes that require coordinated regulation among all
the cells within the developing heart. Clearly, growth factors and
adhesion molecules such as the integrins play a role in shaping the
heart (Kim et al. 1999); however, the presence of the syndecans
suggests that they are active participants in cardiac morphogenesis
as well.

Proteoglycans are common constituents present within the
developing heart (Handler et al. 1997; Litwack et al. 1998; Zanin et
al. 1999) including syndecan-2 in avian and rodent cardiac tissue.
Northern blot analysis has been used to show that late embryonic
stage chick (Chen et al. 2002) and rat hearts (Asundi et al. 1997)
contain syndecan-2 mRNA. The latter study extended these
findings by suggesting that cardiomyocytes possess the potential to
express syndecan-2. Cultures of cardiomyocytes were found to
express low levels of syndecan-2 mRNA. While studies such as
these demonstrate the ability of cardiac tissue to express syndecan-
2, the pattern of syndecan-2 expression during early stages of
cardiac morphogenesis has not been elucidated.

This study seeks to describe the pattern of expression within the
developing chick heart using a polyclonal antibody shown to

GORDON & LANGFORD

279

recognize chicken syndecan-2 (Chen et al. 2002). The data
presented here demonstrate that syndecan-2 is expressed during
Hamburger and Hamilton (HH) stage 15-25 within the developing
myocardium of the heart. Earlier stages of HH 12-14 showed no
detectable syndecan-2. Additionally, during no stage assayed was
syndecan-2 detectable within the endothelium or the cardiac
mesenchyme. These findings suggest that syndecan-2, expressed
during these early stages of cardiac morphogenesis, may play a role
in cardiac development mediated through growth factor attenua¬
tion, cellular adhesion, extracellular matrix organization or another
cellular activity mediated by a member of the syndecan family.

Materials And Methods

Collection of embryonic chick tissues -VQViiXizQ A chicken eggs
were incubated for variable periods of time at 38°C in a forced air
incubator with a humidified atmosphere. Embryos were staged
according to the method of Hamburger and Hamilton (HH)
(Hamburger & Hamilton 1992). Hearts from appropriately staged
embryos were removed and placed in cold PBS (pH 7.2) and
prepared for histological processing.

Tissue fixation and immunohistochemis try. -Embryonic chick
hearts (HH stage 12 - 25) were frozen at -196°C in liquid nitrogen
cooled 1,1,1,2-tetrafluoroethane (R-134a; Kitten et al. 1987) and
transferred to vials containing frozen n,n-dimethylformamide. The
water content of the tissue was gradually substituted with n,n-
dimethylformamide by increasing the temperature to 5°C over 48
hours. The samples, once equilibrated at room temperature, were
routinely processed through a series of graded ethanol solutions
(25% - 100%), infiltrated with the transition solvent Safeclear
(Fisher Scientific, Pittsburg, PA), embedded in paraffin and
sectioned at a thickness of 5 pm.

After routine removal of the paraffin and rehydration of the
tissue sections, this tissue was incubated in a blocking solution of
3% Carnation Instant Milk in PBS (pH 7.2) at room temperature for
one hour. Primary and secondary antibodies were each diluted in a

280

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

buffer of PBS containing 3% Carnation Instant Milk (pH 7.2). The
primary antibody, (R1891, obtained from Dr. Anne Woods; Chen et
al. 2002) was used at a dilution of 1:500. Sections were then
incubated with a FITC conjugated donkey anti-rabbit IgG (Jackson
Immunologicals, West Grove, PA) at a dilution of 1:100. Sections
were examined using an Olympus BX50F compound light
microscope equipped with ultraviolet epi-illumination. Images
were photographed using a Nikon Coolpix 5000 digital camera.

Results

During early stages of cardiac development, the heart begins as a
muscular tube (i.e., myocardium) with an internal lining of
endothelial cells (i.e., endocardium). An acellular milieu of
proteins and carbohydrates, forming the cardiac jelly, separates the
endothelium from the surrounding layer of myocardial cells.
Throughout these stages (HH 12-14), the heart is being transformed
from this “simple tube” into a multi-chambered organ via a process
defined as looping. This dextral curvature of the heart initially
creates two enlarged areas of the heart: the primitive ventricle and
smaller, primitive atrium. Neither the myocardium nor the
endocardium exhibit any detectable syndecan-2 staining (Figs, la &
b).

As cardiogenesis proceeds through HH 15-16, the myocardial
layer begins to expresses syndecan-2 in detectable levels (Figs. Ic
& d). Myocardial cells in all regions of the developing heart
exhibited weak staining for syndecan-2 suggesting that there is no
regional difference in the expression of this proteoglycan during
cardiac development. Interestingly, the outer surface of the
myocardium consistently expresses syndecan-2 (Fig. Id). During
these stages, the epicardium begins to invest the outer surface of the
heart. Thus, the localization of syndecan-2 to the myocardial
surface suggests that this proteoglycan may play a role in epicardial
cell migration. Also, the endocardium continues to lack detectable
syndecan-2 (Fig. Id, arrowhead).

GORDON & LANGFORD

281

Figure 1. Immunohistochemical analysis of syndecan-2 expression in early chick
embryonic cardiac tissue. Syndecan-2 is not detected in the endothelium or
myocardium of a HH stage 12 heart (a, b). The myocardium begins to exhibit
syndecan-2 expression through HH stage 16 (b, c). The bright oval-shaped
structures are auto-fluorescent chicken red blood cells (c-d). All sections are
sagittal sections. Arrow - syndecan-2 positive staining in the outer myocardial
layer; arrowhead - endothelium; M - myocardium; At - atrium; V - ventricle.
Scale Bars = 200|im (a-c); lOOpm (d).

During the stages HH 18-25, syndeean-2 beeomes inereasingly
expressed throughout the myoeardium (Fig. 2). Syndecan-2
expression continues to show no regional differences within the
myocardium (i.e., atrium, ventricle, or outflow tract). In HH 18
hearts, a punctate staining pattern for syndecan-2 becomes evident
throughout the heart including the ventricle (Fig. 2a) and outflow
tract (Fig. 2b). This pattern is consistent with that observed on the
surface of cultured fibroblasts using the same antibody (Chen et al.
2002). As cardiac morphogenesis proceeds through stage HH 21,
valvulogenesis and septation begins and thus, an epithelial -
mesenchymal transition occurs within the endocardial cushions of
the atrioventricular region and the outflow tract. No detectable
syndecan-2 correlated with cardiac mesenchyme in the atrio-

282

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

Figure 2. FIH stage 18 (a, b) and 21 (c, d) chick myocardium continues to express
syndecan-2. Ventricle (a, d); Outflow tract (b); and Atrioventricular cushion
tissue (c). The bright oval-shaped structures are auto-fluorescent chicken red
blood cells (a-d). All sections are sagittal sections. Arrow - syndecan-2 staining
in the myocardial layer; arrowhead - endothelium; Av - atrioventricular canal; V
- ventricle; Ot - outflow tract; asterisk - cardiac mesenchyme. Scale bar =
100pm and applies to a-d.

ventricular cushions of stage HH 21 hearts (Fig. 2c; asterisk).
Additionally, the endocardium lacked detectable levels of
syndecan-2 (Figs. 2b & c).

The pattern of syndecan-2 expression, observed in earlier stages,
continues and is more prominent in HH 25 hearts (Fig. 3).
Syndecan-2 is heavily expressed in the myocardial layers
throughout the heart (Figs. 3a - 3d). While the staining is denser in
these stages, the punctate nature of the expression pattern remains.
In these later stages, cardiac morphogenesis has continued and
through the epithelial-mesenchymal transition has fully populated
the endocardial cushions with mesenchyme. However, no detecta-

GORDON & LANGFORD

283

Figure 3. HH stage 25 chick myocardium continues to express syndecan-2. In this
coronal section, the heart exhibits no apparent regional difference in myocardial
expression of syndecan-2: outflow tract myocardium (a, c); atrial myocardium
(b); ventricular myocardium (c); or the myocardium of the atrioventricular region
(d). Syndecan-2 was not detected in cardiac mesenchymal tissue (asterisk) in
either the atrioventricular region (d) or outflow tract (a, c). HH 25 heart lacking
primary antibody was used as a negative control (f). Only auto-fluorescent red
blood cells are visible in the negative control. Low magnification of HH stage 25
heart (e) is included as reference for high magnification micrographs of the same
section (a-d, f). Adjacent sections were stained for syndecan-2 (d) or from a
negative control (f). Arrow - syndecan-2 staining in the myocardial layer; At -
atrium; Av - atrioventricular canal; V - ventricle; Ot - outflow tract; asterisk -
cardiac mesenchyme. Scale bars = 200pm and applies to a-d, f; 600pm for e.

ble syndecan-2 was observed in HH 25 endocardial cushion tissue
(Figs. 3a, 3c - 3d).

284

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

Discussion

Using an antibody shown to be specific for chicken syndecan-
2 core protein (Chen et al. 2002), these data demonstrate that
syndecan-2 begins to be expressed by cardiomyocytes
immediately prior to the formation of the epicardium. These
results are consistent with previous Northern blot analysis studies
that demonstrate syndecan-2 mRNA production by cardiac tissue.
Chick hearts from day 9-15 were shown to produce syndecan-2
mRNA throughout this developmental period (Chen et al. 2002).
Syndecan-2 mRNA also was detected in late stage (day 18) rat
hearts, although at much lower levels than that of syndecan-3 and
glypican (Asundi et al. 1997). Using isolated and cultured
cardiomyocytes or cardiac non-myocytes, syndecan-3 was shown
to be expressed by the later while syndecan-2 was expressed in
cells from both cultures. While syndecan-2 expression in cardiac
non-myocytes is in apparent contrast to the strict cardiomyocyte
expression observed in the current study, a direct comparison of
data obtained from an immunohistochemical analysis of early
stage chick cardiac tissue (day 3-5) with molecular analysis of
later stage rat cardiac mRNA is difficult due to the species
difference and the tremendous morphogenetic activity that occurs
between early and late stages of embryogenesis. Additionally,
syndecans have demonstrated a tremendous ability to be spatially
and temporally regulated during development. Syndecan-2, in
particular, has been suggested to be the predominant syndecan
expressed during development (Tkachenko et al. 2005) and has
shown alteration in its expression pattern in response to other
syndecan isoforms or other heparan sulfate proteoglycans as
embryogenesis proceeds. This is exemplified by the pattern of
expression of syndecan-4 in cultured cardiomyocytes. In cultures
of neonatal cardiomyocytes, syndecan-4 localizes to the
perinuclear region; however, it is expressed in focal adhesions in
cultures of adult cardiomyocytes (VanWinkle et al. 2002). Thus,
syndecan-2 likely is expressed early in cardiac morphogenesis

GORDON & LANGFORD

285

and possibly downregulated later in embryogenesis as other
syndecan isoforms are expressed in cardiac tissue.

During cardiac morphogenesis, regulation of cell proliferation,
extracellular matrix deposition and organization are critical
biological activities that occur and may be potentiated by
members of the syndecan family. Fibroblast growth factor has
long been known to modulate cardiac morphogenesis and more
recently to regulate cardiomyocyte differentiation (Sheikh et al.
2004; Rosenblatt- Velin et al. 2005; Kruithof et al. 2006). This
potent mitogenic factor requires the presence of heparan sulfate
(as found on syndecans) as coreceptors for FGF receptor binding
and signaling (Bemfield & Hooper 1991; Mundhenke et al.
2002). Transforming growth factor beta (TGFp) is still another
potent regulator of cellular activity that may regulate cardiac
morphogenesis (Ross et al. 1993). Recent studies have
demonstrated the influence of TGFp on cardiomyocyte
differentiation (Sheikh et al. 2004; McKoy et al. 2007).
Interestingly, syndecan-2 has been demonstrated to be required
for TGFp signaling in fibroblasts (Chen et al. 2004). Thus,
syndecan-2 may be functioning in the heart to regulate TGFp
mediated signaling of cardiomyocytes in vivo. Yet still another
factor demonstrated to play a role in shaping the developing heart
is bone morphogenetic protein (BMP) (Somi et al. 2004a).
Similar to the other factors discussed, the activity of BMP is
modulated by syndecans (Fisher et al. 2006) and appears to
regulate cardiomyocytes activity (Somi et al. 2004b; Sugi et al.
2004; Ma et al. 2005; Kruithof et al. 2006).

Another role for syndecan-2 within the myocardium may be
that of a matrix organizer. The cardiac jelly lying between the
myocardium and endocardium is largely synthesized by
myocardial cells. These cells not only produce many of the
morphogens that will lead to the epithelial-mesenchymal-
transition in the endocardial cushions, but also produce many of

286

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

the matrix components that the mesenchymal cells will utilize as
a migratory substratum. In addition to the collagens present,
fibronectin and laminin are also found within the endocardial
extracellular matrix (Bouchey et al. 1996; Nakajima et al. 1997).
Recently, syndecan-2 has been shown to organize both of these
adhesion molecules into a functional fibrillar network (Klass et
at. 2000). In addition to organizing matrix components,
syndecans facilitate cellular adhesion to matrix components.
Studies of syndecan-4 have provided the bulk of evidence for the
role of syndecans in mediating cell adhesion. The ability of
fibroblasts to spread on fibronectin have been shown to require
integrins and syndecan-4 (Saoncella et al. 1999), while carcinoma
cell spreading on vitronectin requires syndecan-1 (Beauvais et al.
2004). With syndecan-2 as the primary syndecan expressed
during development, this proteoglycan may aid in the adhesion of
cardiomyocytes early in cardiac morphogenesis.

Additional evidence for the role of syndecans in mediating
cellular activity has been attained using models of wound healing
and tissue repair. Increased cellular proliferation in response to
released growth factors, altered cellular adhesion and matrix
remodeling are all processes shared between developmental
events and wound healing in adult tissues. Just as syndecans
have diverse patterns of expression and activities during
development, they have also been Implicated in the mediation of
cellular processes leading to proper wound healing in many tissue
types (Worapamom et al. 2002; Elenius et al. 2004; Fears &
Woods 2006). Specific to cardiac tissue, evidence is
accumulating that suggests syndecans play a role in cardiac repair
following myocardial infarcts. All syndecan isoforms were
shown to be upregulated following a myocardial infarct in mice
(Finsen et al. 2004). Additionally, specific upregulation of
syndecan-1 in the infarct tissue appears to decrease the risk of
dilation and dysfunction in the experimental model system
(Vanhoutte et al. 2007). Taken together, all evidence points to a

GORDON & LANGFORD

287

dynamic role for the syndeean family of proteoglycans during
cardiac morphogenesis and during repair following eardiae
injury. Cardiomyocyte expression of syndecan-2 during early
stages of eardiae morphogenesis illustrates how this proteoglycan
could be faeilitating multiple roles during this eritical period of
embryogenesis.

Acknowledgments

We would like to thank Dr. Anne Woods for her generous gift
of the immunological reagents used in this study.

Literature Cited

Akhurst, R. J., S. A. Lehnert, A. Faissner & E. Duffie. 1990. TGF beta in murine
morphogenetic processes: the early embryo and cardiogenesis. Development,
108:645-656.

Asundi, V. K., B. F. Keister, R. C. Stahl & D. J. Carey. 1997. Developmental
and cell-type-specific expression of cell surface heparan sulfate proteoglycans
in the rat heart. Exp. Cell Res., 230:145-153.

Beauvais, D. M. , B. J. Burbach & A. C. Rapraeger. 2004. The syndeean- 1
ectodomain regulates alphavbeta3 integrin activity in human mammary
carcinoma cells. J. Cell Biol., 167:171-181.

Bemfield, M. & K. C. Hooper. 1991. Possible regulation of FGF activity by
syndeean, an integral membrane heparan sulfate proteoglycan. Ann. N Y
Acad. Sci., 638:182-194.

Bouchey, D., W. S. Argraves & C. D. Little. 1996. Fibulin-1, vitronectin, and
fibronectin expression during avian cardiac valve and septa development.
Anat. Rec., 244:540-551.

Chen, L., J. R. Couchman, J. Smith & A. Woods. 2002. Molecular
characterization of chicken syndecan-2 proteoglycan. Biochem. J., 366(Pt)
2:481-490.

Chen, L., C. Klass & A. Woods. 2004. Syndecan-2 regulates transforming
growth factor-beta signaling. J. Biol. Chem., 279( 1 6): 1 57 1 5- 1 57 1 8.

Corless, C. L., A. Mendoza, T. Collins & J. Lawler. 1992. Colocalization of
thrombospondin and syndeean during murine development. Dev. Dyn.,
193:346-358.

Elenius, V., M. Gotte, O. Reizes, K. Elenius M. Bemfield. 2004. Inhibition by
the soluble syndeean- 1 ectodomains delays wound repair in mice
overexpressing syndecan-l. J. Biol. Chem., 279:41928-41935.

288

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

Fears, C. Y. & A. Woods. 2006. The role of syndecans in disease and wound
healing. Matrix Biol., 25:443-456.

Finsen, A. V., P. R. Woldbaek, J. Li, J. Wu, T. Lyberg, T. Tonnessen & G.
Christensen. 2004. Increased syndecan expression following myocardial
infarction indicates a role in cardiac remodeling. Physiol. Genomics, 16:301-
308.

Fisher, M. C., Y. Li, M. R. Seghatoleslami, C. N. Dealy & R. A. Kosher. 2006.
Heparan sulfate proteoglycans including syndecan-3 modulate BMP activity
during limb cartilage differentiation. Matrix Biol., 25:27-39.

Hamburger, V. & H. L. Hamilton. 1992. A series of normal stages in the
development of the chick embryo. 1951. Dev. Dyn., 195:231-272.

Handler, M., P. D. Yurchenco & R. V. lozzo. 1997. Developmental expression
of perlecan during murine embryogenesis. Dev. Dyn., 210:130-145.

Kim, Y. Y., C. S. Lim, Y. H. Song, J. Ahnn, D. Park & W. K. Song. 1999.
Cellular localization of alpha3betal integrin isofonns in association with
myofibrillogenesis during cardiac myocyte development in culture. Cell
Adhes. Commun., 7:85-97.

Kitten, G. T., R. R. Markwald & D. L. Bolender. 1987. Distribution of basement
membrane antigens in cryopreserved early embryonic hearts. Anat. Rec.,
217(4):379-390.

Klass, C. M., J. R. Couchman & A. Woods. 2000. Control of extracellular matrix
assembly by syndecan-2 proteoglycan. J. Cell Sci., 1 13 ( Pt 3):493-506.

Kruithof, B. P., B. van Wijk, S. Somi, M. Kruithof-de Julio, J. M. Perez Pomares,
F. Weesie, A. Wessels, A. F. Moorman, M. J. van den Hoff 2006. BMP and
FGF regulate the differentiation of multipotential pericardial mesoderm into
the myocardial or epicardial lineage. Dev. Biol., 295:507-522.

Langford, J. K., Y. Yang, T. Kieber-Emmons & R. D. Sanderson. 2005.
Identification of an invasion regulatory domain within the core protein of
syndecan- 1. J. Biol. Chem., 280:3467-3473.

Litwack, E. D., J. K. Ivins, A. Kumbasar, S. Paine-Saunders, C. S. Stipp & A. D.
Lander. 1998. Expression of the heparan sulfate proteoglycan glypican-1 in
the developing rodent. Dev. Dyn., 2 1 1 :72-87.

Lopes, C. C., C. P. Dietrich & H. B. Nader. 2006. Specific structural features of
syndecans and heparan sulfate chains are needed for cell signaling. Braz. J.
Med. Biol. Res., 39:157-167.

Ma, L., M. F. Lu, R. J. Schwartz & J. F. Martin. 2005. Bmp2 is essential for
cardiac cushion epithelial-mesenchymal transition and myocardial patterning.
Development, 132:5601-5611.

McFall, A. J. & A. C. Rapraeger. 1998. Characterization of the high affinity cell¬
binding domain in the cell surface proteoglycan syndecan-4. J. Biol. Chem.,
273:28270-28276.

McKoy, G., K. A. Bicknell, K. Patel & G. Brooks. 2007. Developmental
expression of myostatin in cardiomyocytes and its effect on foetal and
neonatal rat cardiomyocyte proliferation. Cardiovasc. Res., 74:304-312.

GORDON & LANGFORD

289

Mundhenke, C., K. Meyer, S. Drew & A. Friedl. 2002. Heparan sulfate
proteoglycans as regulators of fibroblast growth factor-2 receptor binding in
breast carcinomas. Am. J. Pathol., 160:185-194.

Nakajima, Y., M. Morishima, M. Nakazawa, K. Momma & H. Nakamura. 1997.
Distribution of fibronectin, type I collagen, type IV collagen, and laminin in
the cardiac jelly of the mouse embryonic heart with retinoic acid-induced
complete transposition of the great arteries. Anat. Rec., 249:478-485.

Neuhaus, H., V. Rosen & R. S. Thies. 1999. Heart specific expression of mouse
BMP- 10 a novel member of the TGF-beta superfamily. Mech. Dev., 80:181-
184.

Rapraeger, A.C. 2001. Molecular interactions of syndecans during development.
Semin. Cell Dev. Biol., 12:107-116.

Rosenblatt- Velin, N., M. G. Lepore, C. Cartoni, F. Beermann & T. Pedrazzini.
2005. FGF-2 controls the differentiation of resident cardiac precursors into
functional cardiomyocytes. J. Clin. Invest., 115:1724-1733.

Ross, J., D. R. Janero & D. Hreniuk. 1993. Identification and biochemical
characterization of a heart-muscle cell transforming growth factor beta-1
receptor. Biochem. Pharmacol., 46:51 1-516.

Saoncella, S., F. Echtermeyer, F. Denhez, J. K. Nowlen, D. F. Mosher, S. D.
Robinson, R. O. Hynes & P. F. Goetinck. 1999. Syndecan-4 signals
cooperatively with integrins in a Rho-dependent manner in the assembly of
focal adhesions and actin stress fibers. Proc. Natl. Acad. Sci. USA, 96:2805-
2810.

Sheikh, F., C. J. Hirst, Y. Jin, M. E. Bock, R. R. Fandrich, B. E. Nickel, B. W.
Doble, E. Kardami & P. A. Cattini. 2004. Inhibition of TGFbeta signaling
potentiates the FGF-2-induced stimulation of cardiomyocyte DNA synthesis.
Cardiovasc. Res., 64:516-525.

Somi, S., A. A. Buffing, A. F. Moorman & M. J. Van Den Hoff. 2004a.
Dynamic patterns of expression of BMP isoforms 2, 4, 5, 6, and 7 during
chicken heart development. Anat. Rec. A Discov. Mol. Cell. Evol. Biol.,
279:636-651.

Somi, S., A. A. Buffing, A. F. Moorman & M. J. Van Den Hoff. 2004b.
Expression of bone morphogenetic protein- 10 mRNA during chicken heart
development. Anat. Rec. A Discov. Mol. Cell. Evol., Biol. 279:579-582.

Sugi, Y., H. Yamamura, H. Okagawa, R. R. Markwald. 2004. Bone
morphogenetic protein-2 can mediate myocardial regulation of
atrioventricular cushion mesenchymal cell formation in mice. Dev. Biol.,
269:505-518.

Tkachenko, E., J. M. Rhodes & M. Simons. 2005. Syndecans: new kids on the
signaling block. Circ. Res., 96:488-500.

Vanhoutte, D., M. W. Schellings, M. Gotte, M. Swinnen, V. Herias, M. K. Wild,
D. Vestweber, E. Chorianopoulos, V. Cortes, A. Rigotti, M. A. Stepp, F. Van
de Werf, P. Carmeliet, Y. M. Pinto & S. Heymans. 2007. Increased

290

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

expression of syndeean-1 protects against cardiac dilatation and dysfunction
after myocardial infarction. Circulation, 115:475-482.

VanWinkle, W. B., M. B. Snuggs, E. L. De Hostos, L. M. Buja, A. Woods & J. R.
Couchman. 2002. Localization of the transmembrane proteoglycan
syndecan-4 and its regulatory kinases in costameres of rat cardiomyocytes: a
deconvolution microscopic study. Anat. Rec., 268:38-46.

Worapamorn, W., Y. Xiao, H. Li, W. G. Young & P. M. Bartold. 2002.
Differential expression and distribution of syndecan-1 and -2 in periodontal
wound healing of the rat. J. Periodontal Res., 37:293-299.

Zanin, M. K., J. Bundy, H. Ernst, A. Wessels, S. J. Conway & S. Hoffman. 1999.
Distinct spatial and temporal distributions of aggrecan and versican in the
embryonic chick heart. Anat. Rec., 256:366-380.

Zhang, L., G. David & J. D. Esko. 1995. Repetitive Ser-Gly sequences enhance
heparan sulfate assembly in proteoglycans. J. Biol. Chem., 270:27127-27135.

JKL at: klangford@sfasu.edu

TEXAS J. OF SCI. 59(4):291-300

NOVEMBER, 2007

A KEY TO THE COMMON SEED SHRIMP
(CRUSTACEA: OSTRACODA) OF THE PLAYA LAKES OF THE
LLANO ESTACADO REGION OF NORTHWESTERN TEXAS

Francis R. Horne and Ned E. Strenth

Department of Biology, Texas State University
San Marcos, Texas 78666 and
Department of Biology, Angelo State University
San Angelo, Texas 76909

Abstract.-A taxonomic key to those speeies of seed shrimp (Crustacea:
Ostraeoda) known to occur in the playa lakes of the Llano Estaeado region of the
Southern High Plains of Texas is provided. This ineludes 10 speeies representing
eight genera in five families of these temporary pool crustaceans. Identifieation
utilized in this key is based primarily upon the external eharaeteri sties of the earapace
of mature specimens.

Playa lakes of the Southern High Plains represent one of the
more unique inland aquatic habitats of North America and have
received considerable attention due to both their unusual biological
and physical characteristics (Reeves 1966; Rowell 1971; Home
1974; Osterkamp & Wood 1987; Proctor 1990; Smith 2003;
Haukos & Smith 2004). Due to the temporary nature of these
habitats, they are characterized by periods of intermittent drying
and wetting, and undergo wide fluctuations in physical and
chemical conditions (Reeves 1966; Sublette & Sublette 1967;
MacKay et al. 1990).

Eaunal diversity of several playas of eastern New Mexico and
western Texas has been studied by Sublette & Sublette (1967) and
MacKay et al. (1990). Playas are typically dominated by cmsta-
ceans during the early stages of the recharge cycle (one to four
weeks) and by insects during the later stages. The cmstacean fauna
include members of the Branchiopoda (clam shrimp, fairy shrimp
and tadpole shrimp), Cladocera, Copepoda and Ostraeoda, all of
which have eggs or larval stages that are resistant to desiccation and
extreme temperatures. Because of the temporary and unpredictable
nature of the playa wet phase, cmstacean inhabitants can be

292

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

considered as fugitive species (Hutchinson 1951). Eggs must hatch,
larvae develop, and adults reproduce in a few days or weeks before
the play a dries.

The ostracod component in playas often differ significantly
(Home 1996). Because of their infrequent encounter, small size and
overall similarity in appearance, they are often unreported or
remain unidentified and listed only as “ostracods” or “ostracod
species” (Merickel & Wangberg 1981; MacKay et al. 1990).
However, ostracods often represent an important biotic component
of playas of this region of northwest Texas. Most freshwater
species are cosmopolitan in distribution (Tressler 1947; 1954; 1959;
Ferguson 1967a; 1967b; Havel et al. 1990). Their resistant stages
allow for easy dispersal (Proctor 1964; Proctor & Malone 1965;
Home 1966; 1993).

Descriptions of Texas ostracods are available in references by
Ferguson (1967a; 1967b) and Tressler (1954). Delorme’s numer¬
ous papers (1967; 1969; 1970a; 1970b; 1970c; 1970d) on the fresh¬
water ostracods of Canada also are useful when working with this
faunal group.

Although research emphasis today has tended to shift toward
molecular studies, morphological characters are the link between
extant and fossil species (Gutentag & Benson 1962; Wheeler et al.
2004). Morphological characters are those commonly used in field
studies and in laboratory identifications, and also are the ones used
in this paper.

This study provides playa researchers of the Llano Estacado with a
single listing (Table 1) of ostracod species likely encountered as well
as a taxonomic key based primarily upon the external characteristics
of the carapace of mature specimens. Ten species representing eight
genera in five families of the class Ostracoda are given in the
following key.

HORNE & STRENTH

293

Table 1. Listing of species of ostracods known to occur in the playa lakes of the Llano
Estacado region of northwest Texas,

FAMILY

SPECIES

Cyprididae

Subfamily Cypridinae

Heterocypris antillensis
Megalocypris ingens
Megalocypris pseudoingens
Megalocypris gnathostoma

Subfamily Cypridopsinae

Cypridopsis vidua
Potamocypris smaragdina

Cyclocyprididae

Physocypria globula

Candonidae

Candona patzucaro

Llyocoprididae

Pelocypris tuberculatum

Entocytheridae

Limnocythere sanctipatrici

Study Area

The playa lakes included in this study are located in a 43 county
geographical area bordered on the north by the Canadian River and
on the east and west by escarpments (Osterkamp & Wood 1987).
The identification key presented here is based on specimens from
playas of the counties of Andrews, Crosby, Dawson, Gaines, Garza,
Lubbock, Lynn, Martin and Terry (Home 1996). Following the
classification of pluvial lake basins by Reeve’s (1966), all of the
playas are type VI, except for the saline playas which are type IV.

Use of the Key

Only a dissecting microscope with low magnification is required
for ostracod identification. No dissection or examination of
appendage morphology is necessary for identification to species
level. Scanning electron micrographs are presented here to facili¬
tate and confirm identification (Figure 1).

294

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

HORNE & STRENTH

295

Figure 1. Playa ostracods: (a) Male Candona patzucaro\ (b) Female Candona patzucaro\
(c) Candona patzucaro 7th instar; (d) Heterocypris antillensis: (e) Physocypria
globula; (f) Cypridopsis vidua', (g) Male Limnocythere sanctipatrici; (h) Female
Limnocythere sanctipatrici; (i) Potamocypris smaragdina’, (j) Megalocypris
gnathostoma; (k) Megalocypris ingens’, (1) Megalocypris pseudoingens', (m)
Pelocypris(^Pseudoillocypris) tuberculatum. Scale bars: range from 200 um to 3
mm.

The 7th instar of Candona patzucaro (Fig. Ic) is included in
couplet 4a due to the ability of this stage to undergo desiccation and
therefore appear earlier than other species of ostracods following
the initial filling (or recharge) of the playa habitat. Two of the

296

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

larger ostracods, Megalocypris ingens and Megalocypris
pseudoingens, are very similar in appearanee and size, but oeeur in
different types of play as. Megalocypris ingens oeeurs only in saline
lakes, while M pseudoingens inhabits freshwater playas. Delorme
(1969) eonsidered M pseudoingens as uneommon. Although all of
the speeies have been eolleeted from the ephemeral playas of the
Southern High Plains of Texas (Home 1993; 1996), many also
oeeur in stoek tanks, watering troughs and other permanent bodies
of water. Speeimens are deposited in the Museum of Natural
History (USNM No. 413603).

Key to Species

1 . Lateral surfaee of adult earapaee smooth or nearly smooth (w/o

heavy pits and/or grooves) . 2

la. Lateral surfaee of adult earapaee highly seulptured (heavily or
obviously pitted with one or more grooves) . 9

2. Length of adult earapaee less than 2.0 mm . 3

2a. Length of the adult earapaee greater than 2.0 mm . 7

3. Length of adult earapaee less than 2.0 but greater than 1.0 mm.
. 4

3a. Length of adult earapaee less than 1 .0 mm . 5

4. Carapaee greater than 1.0 mm, but less than 1.7 mm ovate in

sideview; venter eonvex or flat right valve bears small rounded
dentieles. Marginal zone of the anterior, antero ventral and
posteroventral regions of the right valve bear small rounded
dentieles . Heterocypris antillensis (Fig. Id).

HORNE & STRENTH

297

4a. Carapace white, greater than 1.0 mm, but less than 1.7 mm,
ovate in sideview; venter convex or flat, characteristic
posterior-lateral surface, faint, but distinct reticulation, female

with blunt posterior aspects, male more rounded .

. . Candona patzucaro (Figs, la, lb, & Ic).

5. Carapace inflated in dorsal view, tumid, width about equal to

height, subtriangular in side view, surface pitted, shell length
less than 0.8mm . Cypridopsis vidua (Fig. If).

5a. Carapace subovate (disc-like) or crescentric (hatchet-like) in
side view; valves compressed in dorsal view . 6

6. Carapace crescentric/hatchet shaped in side view, surface

setaceous; valve overlap distinctive, shell length less than 0.75
mm . Potamocypris smaragdina (Fig. li).

6a. Carapace subovate or disc-like in side view, surface not

setaceous, shell length less than 0.6 mm .

. Physocypria globula {¥ig. le).

7. Length of shell 3.3 - 3.5 mm, tubular-shaped, subtrapezoidal

anterior and posteriorly rounded, setae along margins except
dorsally . Megalocypris gnathostoma (Fig. Ij).

7a. Length of carapace 3.0 - 4.0 mm, freshwater or saline . 8

8. Shell subtrapezoidal; anterior end rounded, posterior end

slightly rounded to truncated, uncommon, found only in
freshwater . Megalocypris pseudoingens (Fig. 11).

8a. Shell subtrapezoidal; both anterior and posterior ends rounded,

not truncated; restricted to saline playas .

. Megalocypris ingens (Fig. Ik).

298

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

9. Lateral shell surface pitted with distinct surface reticulations,
two median sulci, may contain lateral umbo-shaped
protuberances, oblong 1.4 to 1.8 mm, calcareous distinct
tubercles, spines along anterior and posterior margins,

prominent on the anterodorsal margins of both valves .

. Pelocypris (=Pseudoillocypris) tuberculatum (Fig. Im).

9a. Lateral carapace surface reticulate, valves <0.3 mm, bears
three to four lateral protuberances, contains two median sulci,

may or may not be tuberculate .

. Limnocythere sanctipatrici (Figs. Ig & Ih).

Acknowledgements

Appreciation is expressed to Joe Koke, Texas State University,
San Marcos, Texas, for assisting in preparing scanning electron
micrographs, Vernon W. Proctor, Texas Tech University, Lubbock,
Texas, for his many discussions on playa lakes, and Richard
Forester, USGS, Denver, Colorado, for confirmation and
identification of ostracods to species.

Literature Cited

Delorme, L. D. 1967. Field keys and methods of eolleeting freshwater ostracods in
Canada. C^7«. J. Zoo/., 45:1275-1281.

Delorme, L. D. 1969. On the identity of ostracode genera Cypriconcha and
Megalocypris. Can. J. ZooL, 47:27 1 -28 1 .

Delorme, L. D. 1970a. Freshwater Ostracods of Canada. Part 1. Cypridinae. Can. J.
ZooL, 48:153-168.

Delorme, L. D. 1970b. Freshwater Ostracods of Canada. Part II. Subfamily
Cypridopsinae and Herpatocypridinae, and family Cyclocyprididae. Can. J.
ZooL, 48:253-266.

Delorme, L. D. 1970c. Freshwater Ostracods of Canada. Part III. Family
Candonidae. Can. J. ZooL, 48:1099-1127.

Delorme, L. D. 1970d. Freshwater Ostracods of Canada. Part IV. Families
Ilyocyrididae, Notodromadidae, Darwinulidae, Cytherideidae, and Entocytheri-
dae. Cypridinae. Can. J. ZooL, 48:1251-1259.

Ferguson, E. 1967a. New ostracods from the playa lakes of eastern New Mexico and
western Texas. Am. Midland Naturalist, 78(1):248-251.

HORNE & STRENTH

299

Ferguson, E. 1967b. New ostracods from the playa lakes of eastern New Mexieo
and western Texas. Trans. Amer. Microsc. Soc., 86(3):244-250.

Gutentag, E. D. & R. H. Benson. 1962. Neogene (Plio-Pleistocene) freshwater
ostracods of the Central High Plains. State Geological Survey Kansas, Bulletin,
157, (4): 1-60.

Haukos, D. A. & L. M. Smith. 2004. Plant communities of playa wetlands in the
southern Great Plains. No. 47, 66 pp.

Havel, J. E,, P. D. N. Hebert & D. Delorme. 1990. Genetics of sexual Ostracoda
from a low Arctic site. J. Evol. Biol, 3:65-84.

Home, F. R. 1966. The effects of digestive enzymes on the hatchability of Artemia
salina eggs. Trans. Amer. Microsc. Soc., 75:271-274.

Home, F. R. 1974. Phyllopods of some southern high plains saline lakes.
Southwestern Naturalist, 18(4):475-479.

Home, F. R. 1993. Survivial strategy to escape desiccation in a freshwater ostracod.
Cmstaceana, 65:53-61.

Home, F. R. 1996. Ostracods of Texas playas. Southwestern Naturalist, 41(4):450-
455.

Hutchinson, G. E. 1951. Copepodology for the ornithologist. Ecology, 32:571-577

MacKay, W. P., S. J. Loring, T. M. Frost & W. G. Whitford. 1990. Population
dynamics of a playa community in the Chihuahuan Desert. Southwestern
Naturalist, 35(4):393-402.

Merickel, W. F. & J. K. Wangberg, 1981. Species composition and diversity of
macroinvertebrates in two playa lakes in the southern high plains. Southwestern
Naturalist, 26(1):153-153.

Osterkamp, W. R. & W. W. Wood, 1987. Playa-lake basins of the southern High
Plains of Texas and New Mexico: Part I. Hydrologic, geomorphic and geologic
evidence for their development. Geol. Soc. Amer. Bull., 99:215-223.

Proctor, V. W. 1964. Viability of cmstacean eggs recovered from ducks. Ecology,
45:656-658.

Proctor, V. W. & C. Malone. 1965. Further evidence of the passive dispersal of
small aquatie organisms via the internal tract of birds. Ecology, 46:728-729.

Proctor, V. W. 1990. Characeae of Llano Estacada (Texas and adjacent New
Mexico) playas. J. Biogeog., 17:75-84.

Reeves, C. C., 1966. Pluvial lake basins of West Texas. J. Geol., 74:269-291.

Rowell, C. M., 1971. Vascular plants of the playa lakes of the Texas Panhandle and
South Plains. Southwestern Naturalist, 15(1):407-417.

Smith, L. M., 2003. Playa Lakes of the Great Plains, University of Texas Press, Austin,
Texas, 257 pp.

Sublette, J. E. & J. S. Sublette, 1967. The limnology of playa lakes on the Llano
Estacado, New Mexieo and Texas. Southwestern Naturalist, 12(4):369-406.

Tressler, W. L. 1947. A checklist of the known speeies of North American fresh¬
water ostracoda. Am. Midland Naturalist, 38:698-707.

Tressler, W. L. 1954. Fresh-water ostracods from Texas. J. Wash. Acad. Sci.,
44:138-149.

300

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

Tressler, W. L. 1959. Ostracoda. p. 657-734. in W. T. Edmondson (ed.).
Freshwater Biology. 2nd edition. John Wiley and Sons, New York, U.S.A, 1248

pp.

Wheeler, Q. D., P. H. Raven & E. O. Wilson. 2004. Taxonomy; Impediment or
Expedient. Seience, 305:285.

FRH at: FH01@txstate.edu

TEXAS J. SCI. 59(4), NOVEMBER, 2007

301

GENERAL NOTES

A FORMAL SYNTHESIS OF
(+)-MURICATACIN FROM L-TARTARIC ACID

Jack M. Southard & Nicole L. Sears

Physical & Environmental Sciences, Texas A&M University-Corpus Christi
Corpus Christi, Texas 78412-5801

Muricatacin (1) is an acetogenin that displays cytotoxic activity
towards certain tumor cell lines found in humans (Figure 1). This
lipid-soluble class of anticancer drug was first isolated from the seeds
of the Soursop tree (Annona muricata) and has received considerable
attention in recent years. This paper describes a formal synthesis of
(+)-muricatacin (1) using tartaric acid (2) as a chiral controller. The
key step in this synthesis is the formation of butenolide 3 by way of
intramolecular cyclization of 4a while under acidic conditions. The
synthesis of 5iS'-(15',2-dihydroxy-ethyl)-dihydro-furan-2-one (5), a
formal precursor to 1 required six steps with an overall yield of 11%.

The y-butyrolactone (or furanone) ring system is present in a
number of naturally occurring compounds, many of which are
biologically active (Couladouros et al. 1999 and references cited
therein). Chiral hydroxy lactones have also proven to be effective as
antifeedants (Hidefumi et al. 1984) and some have displayed cytotoxic
properties against human tumor cells (Cave et al. 1997). Muricatacin
(1), or 5-(l-hydroxy-tridecyl)-dihydro-furan-2-one, is a y-
butyrolactone containing a normal 13-carbon side chain with a 2°
alcohol branching at C5 (Rieser et al. 1991). Both enantiomers are
present in the natural source with the (-)-(4R,5R)-enantiomer being
dominant over the (+)-(4iS',5*S)-enantiomer (Saniere et al. 1995; Van
Aar et al.; 1995; Baylon et al. 2000; Raghavan & Joseph 2003;
Popasavin et al. 2003; Bernard et al. 2003; Quinn et al. 2004).

Results and Discussion

Using the methods of Seebach & Hungerbiihler (1981) and
Mukaiyama et al. (1990), butenolide 3 was prepared from

302

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

Figure 1. Molecular Structure of (+)-Muricatacin (1).

O OH

■Vy'

OH O

OH 0^0 —1^ 0^0

HO—'" WOH BnO— WOH BnO— ’CHO

6 7 8

H e

BnO"'''''''Y''''"^ ^

„x

BnO

C02Et ^ W

— V=/ BnO—

COoEt

4a 275-9:1 4b

Overall yield = 1 1 %

No. of steps = 7 J HO OH

HO" ^ ^ (see Saniere 1995)
OH

BnO

COoEt

Figure 2. Synthesis of 5, a formal precursor to 1. Keys: (a) (i) DMP, CH3OH, cat. p-
TsOH, r.t. (71%); (ii) LAH, Et20, (88%); (b) BnBr, THF, NaH, 66‘^C (70%); (c)

(C0C1)2, dimethyl sulfoxide, EtsN, DCM, -78-OT (53%); (d) Ph3P=CHC02Et,
CH3OH, 0°C (Z/£' = 9:1); (e) 3 7VHC1, 25°C; (47% over two steps); (f) H2, = 1.1 atm.,
Pd/C (100%).

commercially available L-tartaric acid (2) as shown in Figure 2.
Conversion of 2 to the corresponding acetonide was accomplished
by treatment with 1 ,2-dimethoxypropane (DMP) in the presence of
catalytic />-toluenesulfonic acid (p-TsOH) in CH3OH. Subsequent
LAH (lithium aluminum hydride) reduction of the carboxylic acid
groups provided the chiral L-threitol derivative 6 in good yield.
Monobenzylation of 6 gave chiral alcohol 7; all spectroscopic data
of 7 was consistent with that of Seebach (1991). Using the method

TEXAS J. SCI. 59(4), NOVEMBER, 2007

303

Swem & Omura (1978) for the oxidation of the unproteeted 1°
aleohol of 7 gave aldehyde 8 which was isolated by Kugelrohr
distillation and used immediately. Wittig olefmation of freshly
distilled 8 using ethoxycarbonyl triphenylphosphorane
(Ph3P=CHC02Et) resulted in formation of chiral alkenes 4a and 4b
in an approximate 9:1 ratio with the desired Z-isomer, i.e. 4a, being
predominant. Without separation the mixture of 4a and 4b were
dissolved in tetrahydrofuran (THF) along with 3 M aqueous HCl
and stirred at ambient temperature for 24 hours. After extraction
with ethyl acetate, butenolide 3 was isolated via crystallization
using ethyl acetate and hexane. Diol 9 was present as a byproduct;
this species was apparently formed from the deprotection of E-
isomer 4b. Catalytic hydrogenation of 3 affected both
debenzylation and the desired partial reduction of the furanone ring
system to produce 5, a formal precursor to 1 (Saniere et al. 1995).
The spectroscopic data obtained for 5 were consistent with that of
the literature values (Saniere et al. 1995).

Conclusion

Beginning with 2 the overall yield of 5 was approximately 11%
over the course of seven steps. As noted earlier the key step in the
overall process was the intramolecular cyclization of 4a to
butenolide 3. Butenolides, such as 3, have proven to be amenable
and versatile to a broad range of well-known reactions (Southard &
Harding 2005). Furthermore this effort shows the utility L-tartaric
acid as a chiral controller in asymmetric synthesis.

Acknowledgements

Funding for this project was provided by the Robert A. Welch
Foundation.

Literature Cited

Baylon, C., G. Prestat, M. P. Heck, & C. Mioskowski. 2000. Lewis Acid-Catalyzed
Regiospecific Opening of Vinyl Epoxides. Tetrahedron Letters, 41:3829-3831.
Bernard, A. M., A. Frongia, P. P. Piras & F. Seed. 2003. Unexpected
Stereochemistry in the Lithium Salt Catalyzed Ring Expansion of Nonracemic

304

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

Oxaspiropentanes. Formal Synthesis of (-)-(4/?,57?)-Muricatacin and the
Pheromone (7?)-Japonilure. Organic Letters, 5:2923-2926.

Cave, A., C. Chaboche, B. Figadere, J. C. Haramange, A. Laurens, J. F. Peyrat, M.
Pinchon, M. Szlosek, J. Cotte-Lafitte & A. M. Quero. 1997. Study of the
Structure-Activity Relationships of the Acetogenin of Annonaceae, Muricatacin
and Analogues, European Journal of Medicinal Chemistry, 32:617-623.

Couladouros, E. & A. Mihou. 1999. A General Synthetic Route Towards y- and 6-
Lactones. Total Asymmetric Synthesis of (-)-Muricatacin and the Mosquito
Oviposition Pheromone (57?,6N)-6-Acetoxy-hexadecanolide. Tetrahedron
Letteres, 40:4861-4862.

Hidefumi, M., Y. Kimura, M. Higuchi, H. Konno, M. Murai & H. Miyoshi. 2006.
Synthesis of {4R,\5R,\6R,2\S)- and (4R,15N,16N,215)-Rollicosin, Squamostolide,
and their Inhibitory Action with Bovine Heart Mitochondrial complex 1. 2006.
Bioinorganic Medicinal Chemistry, 14:31 19-3130.

Mukaiyama, T., K. Suzuki, T. Yamada & F. Tabusa. 1990. 4-0-Benzyl-2,3-0-
isopropylidene-L-threose: A Useful Building Block for Stereoselective Synthesis
of Monosaccharides, Tetrahedron, 46:265-276.

Popsavin, V., S. Grabez, 1. Krstic, M. Popsavin & J. Djokovic. 2003. A Formal
Synthesis of (+)-Muricatacin from D-xylose, Journal of the Serbian Chemical
Society, 68:795-804.

Quinn, K. J., A. Isaacs & R. Arvary. 2004. Concise Total Synthesis of (-)-
Muricatacin by Tandem Ring-Closing/Cross Metathesis, Organic Letters, 6:4143-
4145.

Raghavan, S. & S. Joseph. 2003. The Sulfmyl Moiety as an Intramolecular
Nucleophile. Part 3: Synthesis of (-)-Muricatacin. Tetrahedron Asymmetry,
14:101-105.

Rieser, M. J., J. F. Kozlowski, K. V. Wood & J. L. McLaughlin. 1991. Muricatacin:
A Simple Biologically Active Acetogenin Derivative from the Seeds of Annona
muricata {annonaceae). Tetrahedron Letters, 32:1 137-1 140.

Saniere, M., 1. Charvet, Y. Le Merrer & J. C. Depezay. 1995. Enantiopure
Hydroxylactones from L-ascorbic and D-isoascorbic acids. Part 1. Synthesis of
(-)-Muricatacin. Tetrahedron 51:1653-1662.

Seebach, D. & E. Hungerbiihler. 1981. Chirale elektrophile Synthesebausteine mit
vier verschiedenen funktionellen Gruppen aus Weinsaure, 2, 3- und 3, 4-Epoxy-
butandiolderivate in alien vier stereoisomeren Formen. Helvetica Chimica Acta,
64:687-694.

Southard, J. M. & K. E. Harding. 2005. A Formal Synthesis of Thymine Polyoxin
C. Tetrahedron Asymmetry 16:1845-1854.

Swern, D. & K. Omura. 1978. Oxidation of Alcohols by “Activated” Dimethyl
Sulfoxide. A Preparative, Steric and Mechanistic Study. Tetrahedron 34:1651-
1660.

Van Aar, M., L. Thijs & B. Zwanenburg. 1995. Synthesis of (4R,5i?)-Muricatacin
and its (47^,55)- Analog by Sequential use of the Photo-induced Rearrangement of
Epoxy Diazomethyl Ketones. Tetrahedron, 51:1 1223-1 1234.

JMS at: jack.southard@tamucc.edu

TEXAS J. SCI. 59(4), NOVEMBER, 2007

305

it "k "k "k "k

RE-OCCURRENCE OF THE TROPICAL GREEN MACRO ALGA,
PENICILLUS CAPITATUS (CHLOROPHYTA: BRYOPSIDALES),
IN THE LOWER LAGUNA MADRE OF SOUTH TEXAS

Joseph L. Kowalski*, Donald L. Hockaday, Gilbert H. Boza, Jr.,
and Hudson R. DeYoe*

"^Department of Biology and the Center for Subtropical Studies
The University of Texas - Pan American
1201 West University Drive, Edinburg, Texas 78541 and
Coastal Studies Laboratory and the Center for Subtropical Studies
The University of Texas - Pan American
South Padre Island, Texas 78597

The siphonaceous green alga Penicillus capitatus Lamarek is a
eommon maeroalga of the Carribean and Gulf of Mexieo (Littler &
Littler 2000). This shaving brush-shaped maeroalga was a eommon
inhabitant of the Lower Laguna Madre, Texas (LLM) until about 40
years ago when the population was deeimated following a salinity
deeline in the fall of 1959, another salinity reduction due to
Hurricane Carla in 1961, and finally a severe freeze in 1962
(Sorensen 1963). Since that time, there have been no reports of P.
capitatus in the LLM. During a reconnaissance trip in July 2006, a
small population was found in the southern portion of the LLM.
The species descriptions by Taylor (1960) and Sorensen (1979)
were used in the identification of specimens. Herbarium specimens
have been deposited in the herbaria of The University of Texas-Pan
American Coastal Studies Laboratory (accession number CO 12),
and the University of Texas at Austin Marine Science Institute.

Penicillus capitatus thalli were mostly concentrated on unvege¬
tated, unconsolidated sediment, colloquially termed ’’potholes" or
"blowouts", surrounded by the seagrass Thalassia testudinum in the
LLM (26° 08’ 48.7" N, 097° 11’ 52.0" W), approximately 7.1 km
north of the Brazos-Santiago Pass. Other macroalgae found with P.
capitatus were Caulerpa mexicana Sonder ex. Kutzing and C.
prolifera E. obovata (Eorsskal) Lamouroux. Halimeda incrassata.

306

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

Caulerpa prolifer a, and Codium taylorii were noted as recent
colonizers to the Laguna Madre (Kaldy 1996; DeYoe & Hockaday
2001). Penicillus capitatus, C. prolifera, and C mexicana were
found growing in monotypic stands in an area of the LLM
characterized by clear, nutrient-poor water 1.2 to 1.3 m deep (Kaldy
& Dunton 1997; Kaldy et al. 2004). Salinity during the July 2006
trip was 37 PSU and water temperature was 31° C. The general
area was surveyed and many of these potholes were recently
vegetated by T. testudinum deduced from a rugose appearance of
the potholes (caused by dead bundle sheaths) and the presence of
dead rhizome material in the sediment. Many of these pothole P.
capitatus plants were found with their holdfasts attached to the
bundle sheath material. On this trip P. capitatus biomass samples
were collected in three 9 cm diameter cores (0.006 m^) driven 10
cm into the seabed. Sediment was washed from the samples
through a 1 mm mesh sieve immediately upon return to the lab and
stored in sealed plastic bags and refrigerated until processed.
Processed samples were dried at 80°C to a constant dry weight and
weighed to the nearest 0.001 g, then ashed in a muffle furnace at
500°C and reweighed to the nearest 0.001 g. Mean biomass (dry
weight) for P. capitatus was 158.03 g m'^ {SE = 11.53 g m"^) and
AFDW 54.47 g {SE = 5.78 g m’^. Sorensen (1963) reported a
mean dry weight of 5.14 g per 4 inch (10.16 cm) diameter core
which is equivalent to 634 g m'^.

A second trip to the same area was made on 3 August 2006 to
characterize the extent and distribution of P. capitatus at the site.
Three 150 m transects, each separated by 50 m and oriented south
to north, were selected that intersected at least one bare area. Along
each transect, every 10 to 15 m, two quadrats 25 cm by 25 cm
(0.0625 m^) were randomly dropped from each side of the boat.
Each quadrat was scored for the presence or absence of P. capitatus
or seagrass and thalli or shoots counted. Approximately one-half of
the 56 sampling points along all transects were judged to be
predominately bare or sparsely covered by seagrass and one-half
were judged to be predominately grassy areas. To test the null

TEXAS J. SCI. 59(4), NOVEMBER, 2007

307

hypothesis that there was no difference in the occurrence of P.
capitatus as influenced by bottom type (vegetated vs. unvegetated
sediments), quadrat counts were examined using a Chi-square test
category one 2x2 contingency table (Zar 1999). Alpha was set at

O. 05. Quadrat counts of thalli and seagrass density were also
converted to a per meter square basis for the purpose of discussion.

From the 56 total quadrat counts made from all three transects,

P. capitatus was found within five quadrats, all within potholes

with a mean density of 9.14 thalli {SE = 4.85, n = 56) and
occurred independent of vegetated sediments (X^ = 9.08, n = 56, \
df). Although no P. capitatus were recorded among seagrass shoots
within given transects, some were noted among Thalassia
testudinum shoots between transects. Mean T. testudinum shoot
density across all transects was 510 shoots m' {SE = 45.03, = 28).

This value is low compared to values found by Herzka & Dunton
(1997) and Kaldy & Dunton (2000) at nearby sites at similar depth
and season. Seagrass shoot densities encountered here may be
influenced by the proximity to the adjacent potholes, especially
since pothole formation may be directly related to smothering by
drift algae, as suspected in the study area. Drift algal mats were
observed overlying seagrass shoots followed by seagrass denuda¬
tion, and resulting short-shoot stubble. If there is a positive
relationship between pothole formation and P. capitatus
recruitment, the result may be an increase in sea bottom available
for algal colonization. Without directly stating such, Sorensen
(1963) implied that greatest thalli densities (several hundred m’^)
occurred on unvegetated sediments. Two potholes were selected in
the study area where Penicillus density appeared uniformly
distributed and counted 30 and 33 thalli per quadrat. This is
equivalent to 480 and 528 thalli m'^, respeetively. Sorensen (1963)
also noted that P. capitatus was less abundant within seagrass beds,
similar to observations made during this study. In subsequent trips
to other LLM research sites, P. capitatus was not found.

308

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

Penicillus capitatus is a common rhizophytic alga, found
primarily in shallow tropical embayments associated with coral and
seagrass ecosystems. However, it is also found in shallow
subtropical regions (Zieman & Zieman 1989). In addition to its
role in ecosystem primary productivity, it and its relatives (Udotea
sp., Acetabularia sp., Halimeda sp.) can be responsible for much of
the calcium carbonate mud deposited in back-reef areas when
present in abundance. To the author’s knowledge, the LLM is the
northernmost extent of the P. capitatus range in the western Gulf of
Mexico. Its present day occurrence in the LLM likely results from
recent introduction from Mexico combined with several years of
moderate winter temperatures and moderate salinities. The same
conclusion was reached by DeYoe & Hockaday (2001) for the
recent records of C. prolifera F. obovata and C. taylorii.

Thorhaug et al. (1971) reported vigorous growth in the
laboratory between 15 and 31.5°C, but inability of plants to survive
more than nine days below 14 and above 34°C. Field studies from
Biscayne Bay, Florida have documented a lower thermal tolerance
of 12 to 14°C for P. capitatus by Biber (2002). Long-term daily
integrated water column temperature data from two stations in the
LLM (Port Isabel and South Padre Island Coast Guard Station)
shows that the minimum winter temperatures were 7.10 (January
1997 - Port Isabel) and 6.30°C (December 1998 - Coast Guard
Station) (DNR, 2006). Low water column temperature values over
the past ten years in the LLM lasted for less than a week and if P.
capitatus had already re-established itself between 1997 and 1998 it
appears that the lowest LLM temperatures were insufficient in
either duration or magnitude to eliminate it from the area again.
Highest water column temperatures in the LLM over the same
period were 34.1 and 34.4°C, appears to be within the upper
temperature tolerance of P. capitatus. For the period between
September 1994 and February 1998 at a mid-bay location, the
lowest salinity was 20.2 PSU during December 1995 (DNR, 2006).
Sustained low temperatures and salinity could adversely affect this

TEXAS J. SCI. 59(4), NOVEMBER, 2007

309

species (Sorensen 1963) but seemed not to have occurred in the past
10 years.

The re-occurrence of P. capitatus in the LLM is notable for two
reasons. First, experimental work has shown competitive inter¬
actions exist between seagrasses and rhizophytic macroalgae.
These include interactions between calcareous green plants, such as
H. incrassata and T. testudinum (Davis & Fourqurean 2001) and C.
prolifera and Halodule wrightii, shoalgrass (Taplin et al. 2005;
Stafford & Bell 2006). Significant increases in the distribution and
abundance of rhizophytic macroalgae in the LLM has the potential
to lead to edaphic resource competition between newly invading
algae and seagrasses. With little to no root/rhizome fraction to
support, rhizophytic algae may have a competitive edge in the LLM
where sediment porewater nitrogen and phosphorus pools are often
half that of seagrass meadows elsewhere along the Texas coast
(Kowalski 1999). Second, calcareous macroalgal species, such as
Halimeda and P. capitatus, are capable of sufficient lime mud
production to influence porewater nutrient levels. Phosphate ions
readily and quickly adsorb to calcium carbonate (aragonite and
calcite) and become unavailable for plant uptake (Short et al. 1990).
Furthermore, P. capitatus secretes the aragonite form of calcium
carbonate which has 20% more surface area than calcite. Davis &
Fourqurean (2001) report that 80% of total plant production by
Halimeda in Florida Bay was from inorganic carbonate at a rate of
about 225g m'^ yeaf\ Phosphorus limits seagrass growth in
Florida Bay, ranging from severe (N to P ratio of 96:1) to moderate
(N to P ratio of 63:1) (Armitage et al. 2005) possibly due to high
levels of aragonite from Halimeda.

The Laguna Madre of Texas has not been as well-studied as
other Texas bays or estuaries. It is yet unclear how and why
recently discovered (and re-discovered) benthic macroalgae have
come to occupy the LLM. The species found recently {H.
incrassata, C. prolifera, C. taylorii and P. capitatus) are benthic
and distinctive. It is very possible that other more cryptic species

310

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

have invaded the LLM from the south but have not yet been
discovered. The possible impacts of these recent algal discoveries
in the LLM ecosystem deserve study.

Acknowledgements

We thank David Kowalski, who discovered the first thallus of
Penicillus, and Chris Pederson for very able field assistance.
Comments by Dr. C. O. Patterson improved the manuscript. This is
contribution CSS 07-002 of the Center for Subtropical Studies.

Literature Cited

Armitage, A. R., T. A. Frankovich, K. L. Heck, Jr, & J. W. Fourqurean. 2005.
Experimental nutrient enrichment causes complex changes in seagrass,
microalgae, and macroalgae community structure in Florida Bay. Estuaries,
28(3):422-434.

Biber, P. 2002. The effects of environmental stressors on the dynamics of three
functional groups of algae in Thalassia testudinum habitats of Biscayne Bay,
Florida, A modeling approach. Unpublished Ph.D. Dissertation, Univ. Miami,
Coral Gables, Florida, 367 pp.

Davis, B. C. & J. W. Fourqurean. 2001. Competition between the tropical alga,
Halimeda incrassata, and the seagrass, Thalassia testudinum. Aq. Bot.,
71(3):217-232.

Division of Nearshore Research. 2006. TCOON data base
(http ://lighthouse .tamucc . edu/TCOON/HomePage) .

DeYoe, H. R. & Donald L. Hockaday. 2001. Range extensions of the seaweeds
Codium taylorii and Caulerpa prolifera into the Lower Laguna Madre of Texas.
Tex. J. Sci., 53(2): 190-192.

Herzka, S. Z. & K. H. Dunton. 1997. Seasonal photosynthetic patterns of the
seagrass Thalassia testudinum in the western Gulf of Mexico. Mar. Ecol. Prog.
Ser., 152:103-117.

Kaldy, J. E. 1996. Range extension of Halimeda incrassata (Chlorophyta,
Bryopsidales): occurrence in the Lower Laguna Madre of Texas. Southwest.
Nat.,41(4):419-423.

Kaldy, J. E. & K. H. Dunton. 2000. Above- and below-ground production, biomass
and reproductive ecology of Thalassia testudinum (turtle grass) in a subtropical
coastal lagoon. Mar. Ecol. Prog. Ser., 193:271-283.

Kaldy, J. E., K. H. Dunton, J. L. Kowalski & Kun-S. Lee. 2004. Evaluation of
environmental factors controlling the success of seagrass revegetation onto
dredged material deposits: a case study in the Lower Laguna Madre, Texas. J.
Coast.Res.,20(l):292-300.

TEXAS J. SCI. 59(4), NOVEMBER, 2007

311

Kowalski, J. L. 1999. Production of the subtropical seagrass, Halodule wrightii
Aschers., in Lower Laguna Madre, Texas. Unpublished MS Thesis, Univ. Tex. -
Pan American, Edinburg, Texas, 105 pp.

Littler, D. S. & M. M. Littler. 2000. Caribbean Reef Plants. Offshore Graphics, Inc.
Washington, D.C., 542 pp.

Short, F. T., W. C. Dennison & D. G. Capone. 1990. Phosphorus-limited growth of
the subtropical seagrass Syringodium filiforme in carbonate sediments. Mar.
Ecol.Prog. Ser., 62:169-174.

Sorensen, L. O. 1963. Growth and decay of Penicillus capitatus Lamarck in the
Lower Laguna Madre of Texas. Publ. Mar. Sci. Inst., 9:105-1 II.

Sorensen, L. O.. 1979. A Guide to the Seaweeds of South Padre Island, Texas.

Gorsuch Scarisbrick, Publishers, Dubuque, Iowa, 123 pp.

Stafford, N. B. & S. S. Bell. 2006. Space competition between seagrass and
Caulerpa prolifer a (Forskaal) Lamouroux following simulated disturbances in
Lassing Park, FI. J. Exp. Mar. Biol. Ecol., 333(1 ):49-57.

Taplin, K. A., E. A. Irlandi & R. Raves. 2005. Interference between the macroalga
Caulerpa prolifera and the seagrass Halodule wrightii. Aq. Bot., 83(3): 175- 186.

Taylor, W. R. 1960. Marine Algae of the Eastern Tropical and Subtropical Coasts of
the Americas. University of Michigan Press, Ann Arbor Michigan, 460 pp.

Thorhaug, A., T. Devany, J. C. Bauer & S. Pepper. 1971. The effect of temperature
on Penicillus capitatus survival in laboratory and field investigations. J. Phycol.,
7(5):5-6.

Zar, J. H. 1999. Biostatistical Analysis. Second Edition. Prentice Hall, New Jersey,
663 pp.

Zieman, J. C. & R. T. Zieman. 1989. The Ecology of the Seagrass meadows of the
West Coast of Florida: A Community Profile. Volume Biological Report
85(7.25). U. S. Department of the Interior - Fish and Wildlife Service Research
and Development, Washington, D. C., 155 pp.

JLK at: kowalski@utpa.edu

REPRODUCTION IN THE REDBACK COFFEE SNAKE,

NINIA SEBAE (SERPENTES: COLUBRIDAE), FROM
SOUTHERN MEXICO AND CENTRAL AMERICA

Stephen R. Goldberg

Department of Biology, Whittier College
Whittier, California 90608

The redback coffee snake, Ninia sebae frequents lowlands and
premontane slopes from Veracruz and Oaxaca, Mexico, southward

312

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

through Central America to southern Costa Rica where it is known
from 4-1,100 m; it is a secretive, nocturnal leaf litter inhabitant
(Savage 2002). There is anecdotal information on its reproductive
biology in Gaige (1936); Stuart (1948); Alavrez del Toro (1960);
Greene (1975); Campbell (1998); Lee (2000); Solorzano (2004).
The purpose of this paper is to add additional information on the
reproductive biology of N. sebae from a histological examination of
museum specimens from southern Mexico and Central America as
part of an ongoing study of the reproductive cycles of neotropical
snakes.

A sample of adult specimens of 56 N. sebae was examined
(females n = 33, mean snout-vent length [SVL] = 245 mm ± 28
SD, range = 191-305 mm; males n = 23, SVL = 216 mm ± 25 SD,
range = = 160-263 mm) from the herpetology collection of the
Natural History Museum of Los Angeles County (LACM), Los
Angeles, California. This sample included specimens from Mexico
(1 female, 1 male), Guatemala (26 females, 17 males), Costa Rica
(3 females, 2 males) and Honduras (3 females, 3 males). Snakes
were collected 1963-1980 . Counts were made of enlarged ovarian
follicles (> 5 mm length) or oviductal eggs. The left testis and vas
deferens were removed from males and the left ovary was removed
from females for histological examination. Tissues were embedded
in paraffin and sectioned at 5 jum. Slides with tissue sections were
stained with Harris’ hematoxylin followed by eosin counterstain
(Presnell & Schreibman 1997). Histological slides were examined
to determine the stage of the testicular cycle and for the presence of
yolk deposition or corpora lutea. There were too few egg clutches
to compare mean clutch sizes between different countries. The
relationship between female SVL and clutch size was examined by
linear regression analysis. An unpaired Mest was used to compare
mean body sizes (SVL) between males and females. Statistical
tests were performed using Instat, vers 3.0b (Graphpad Software,
San Diego, CA).

TEXAS J. SCI. 59(4), NOVEMBER, 2007

313

Material examined following specimens of Ninia sebae
were examined: COSTA RICA (5). Limon Province LACM
154183, 154184, Guanacaste Province LACM 154186, 154190, San
Jose Province LACM 154188 GUATEMALA (43). Huehuete-
nango Department LACM 40056, 40058-40066, 40068-40070,
40073-40076, 40078-40081, 40083-40087, 40090-40096, 40098,
40100-40106, Izabel Department LACM 59119, 59120

HONDURAS (6). Cortes Department LACM 45275 45336 45340,
45341, Copan Department LACM 45388, 45389; MEXICO (2).
Veracruz State LACM 103629, 134846.

The testicular histology of N. sebae was similar to that reported
by Goldberg (2004a) for Ninia maculata. All testes examined
exhibited spermiogenesis with metamorphosing spermatids and
sperm present. Vasa deferentia contained sperm. The following
numbers of males were undergoing spermiogenesis by month:
February (1); March (3); April (15); May (1); July (1); September
(1); December (1). The smallest male to undergo spermiogenesis
measured 160 mm SVL (LACM 134846) and was from December.
Even with no samples from several months, it is apparent N. sebae
has a prolonged period of spermiogenesis that encompasses most, if
not the entire year. Prolonged periods of spermiogenesis have been
reported for other species of snakes from Central America
(Goldberg 2003a; 2003b; Goldberg 2004a; 2004b; 2004c; Goldberg
2006a; 2006b; Goldberg 2007a; 2007b; 2007c; Goldberg 2007d;
Goldberg 2007e; Goldberg 2008).

Females were significantly larger (SVL) than males {df= 54, t =
4.05, P = 0.0002). The monthly changes in the ovarian cycle of N.
sebae are summarized in Table 1. Mean clutch size for 23 females
was 3.2 ± 0.83, range = 2-5. Linear regression analysis revealed a
significant positive correlation between female body size (SVL) in
mm and clutch size for 23 N. sebae females: Y == - 1.55 + 0.02 X, r
= 0.53, P = 0.0091. The smallest reproductively active female
(oviductal eggs) measured 208 mm SVL (LACM 45336) and was
from Honduras. Other records of N. sebae clutches are in Table 2.

314

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

Table 1. Monthly distribution of stages in the ovarian eycle of 33 Ninia sebae. Values
shown are the numbers of females exhibiting each of the four conditions.

Month

No yolk
deposition

Early yolk
deposition

Enlarged follicles
> 12 mm length

Oviductal

eggs

Corpora
lutea only

Mar.

Apr.

July

Aug.

Sept.

Nov.

Table 2. Clutch sizes of Ninbia sebae from the literature.

Clutch Size

Locality

Date

Source

not given

Yucatan, Mexico

July

Gaige 1936

not given

Alta Verapaz, Guatemala

March-June

Stuart 1948

2-4

Chiapas, Mexico

August

Burger & Werler 1954

3 or 4

Chiapas, Mexico

not given

Alvarez del Toro 1960

Atlantida, Honduras

21 July

Meyer 1966

1-3

Veracruz, Mexico

28 August-6 September Greene 1975

Belize

January

Bohuslavek 1996

1-4

Peten, Guatemala

March-September

Campbell 1998

2-4

Yucatan, Mexico

summer

Lee 2000

up to 4

Costa Rica

not given

Solorzano 2004

Two females from Honduras, each with clutches of five eggs
(LACM 45275, 45388) are new maximum clutch sizes for A. sebae.
Three females from Guatemala with oviductal eggs (LACM 40062,
40063, 40080) exhibited concurrent yolk deposition for a
subsequent clutch indicating N. sebae likely produces more than
one clutch per year.

At this point it appears that prolonged periods of spermiogenesis
may be typical for Central American snakes. Also, in view of the
prolonged period of reproductive activity, production of multiple

TEXAS J. SCI. 59(4), NOVEMBER, 2007

315

clutches may also be common. More work is needed before the
variations in the reproductive cycles of Central American snakes
can be ascertained.

Acknowledgments

I thank Christine Thacker (LACM) for permission to examine
specimens Dustin Goto (Whittier College) assisted with histology.
The snakes from Costa Rica are part of the Costa Rica Expeditions
(CRE) collection donated to LACM by Jay Savage.

Literature Cited

Alvarez del Toro, M. 1960. Reptiles de Chiapas. Primera Ed., Institute
Zoologico del Estado, Tuxtla Gutierrez, Chiapas, 204 pp.

Bohuslavek, J. 1996. Ninia sebae sebae (Red Coffee Snake). Reproduction.
Herpetol. Rev., 27:146.

Burger, W. L. & J. E. Werler. 1954. The subspecies of the ring-necked coffee
snake, Ninia diademata, and a short biological and taxonomic account of the
genus. Univ. Kansas Sci. Bull., 36:643-672.

Campbell, J. A. 1998. Amphibians and reptiles of northern Guatemala, the
Yucatan, and Belize. University of Oklahoma Press, Norman, xix + 380 pp.
Gaige, H. T. 1936. 18. Some reptiles and amphibians from Yucatan and

Campeche, Mexico. Carnegie Inst. Wash., 457:289-304.

Goldberg, S. R. 2003a. Reproduction in the speckled racer, Drymobius
margaritiferus (Serpentes: Colubridae), from Mexico and Central America.
Texas J. Sci., 55(3): 195-200.

Goldberg, S. R. 2003b. Reproduction in four species of Dendrophidion from
Costa Rica (Serpentes: Colubridae). Trans. Illinois State Acad. Sci., 96:295-
300.

Goldberg, S. R. 2004a. Reproduction in the coffee snake, Ninia maculata
(Serpentes: Colubridae), from Costa Rica. Texas J. Sci., 56(l):81-84.
Goldberg, S. R. 2004b. Notes on reproduction in the false coral snakes,
Erythrolamprus bizona and Erythrolamprus mimus (Serpentes: Colubridae)
from Costa Rica. Tex. J. Sci., 56(2): 171-174.

Goldberg, S. R. 2004c. Notes on reproduction in the Central American coral
snake, Micrurus nigrocinctus (Serpentes: Elapidae) from Costa Rica. Carib. J.
Sci., 40:420-422.

Goldberg, S. R. 2006a. Note on the testicular cycle of the Costa Rica water
snake, Hydromorphus concolor (Serpentes: Colubridae). Bull. Maryland
Herpetol. Soc. 42:169-170.

316

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

Goldberg, S. R. 2006b. Reproductive cycle of the salmon-bellied racer,
Mastigodryas melanolomus (Serpentes, Colubridae), from Costa Rica.
Phyllomedusa J. Neotrop. HerpetoL, 5:135-138.

Goldberg, S. R. 2007a. Note on the testicular cycle of Godman’s earth snake,
Geophis godmani (Serpentes: Colubridae) from Costa Rica. Bull. Chicago
Herp. Soc., 42:7-8.

Goldberg, S. R. 2007b. Note on the testicular cycle of the northern scorpion-
eater, Stenorrhina freminvillii (Serpentes: Colubridae) from Central America.
Bull. Chicago Herp. Soc., 42:90-91.

Goldberg, S. R. 2007c. Coniophanes fissidens (Brown Spotbelly).

Reproduction. HerpetoL Rev., 38:339.

Goldberg, S. R. 2007d. Note on reproduction of whipsnakes. Genus Chironius
(Serpentes: Colubridae), from Costa Rica. Bull. Chicago Herp. Soc., 42:148-
149.

Goldberg, S. R. 2007e. Notes on reproduction of the adorned graceful brown
snake, Rhadinea decorata (Serpentes, Colubridae), from Costa Rica.
Phyllomedusa J. Neotrop. HerpetoL, 6:151-153.

Goldberg, S. R. 2008. Reproduction in the false fer-de-lance, Xenodon
rabdocephalus (Serpentes: Colubridae) from Costa Rica. The HerpetoL Bull.,
Vol. 103 (in press).

Greene, H. W. 1975. Ecological observations on the red coffee snake, Ninia
sebae, in southern Veracruz, Mexico. Amer. Midi. Nat., 93:478-484.

Lee, J. C. 2000. A field guide to the amphibians and reptiles of the Maya world.
The lowlands of Mexico, Northern Guatemala, and Belize. Comstock
Publishing Associates, Cornell University Press, Ithaca, Viii + 402 pp.

Meyer, J. R. 1966. Records and observations on some amphibians and reptiles
from Honduras. Herpetologica, 22: 1 72- 181.

Presnell, J. K. & M. P. Schreibman. 1997. Humason’s animal tissue techniques,
5^*^ Ed., The Johns Hopkins Press, Baltimore, xi + 572 pp.

Savage, J. M. 2002. The amphibians and reptiles of Costa Rica. A herpetofauna
between two continents, between two seas. University of Chicago Press,
Chieago, Illinois, 934 pp.

Solorzano, A. 2004. Snakes of Costa Rica. Distribution, taxonomy, and natural
history. Instit. Nacional de Biodiversidad, Editorial INBio, Costa Rica, 791

pp.

Stuart, L. C. 1948. The amphibians and reptiles of Alta Verapaz Guatemala.
Misc. Publ. Mus. ZooL, Univ. Mich., 69:1-109.

SRG at: sgoldberg@whittier.edu

"k -k "k -k "k

TEXAS J. SCI. 59(4), NOVEMBER, 2007

317

POSSIBLE VERTEBRATE BURROWS FROM
THE MIOCENE FLEMING FORMATION NEAR HUNTSVILLE,
WALKER COUNTY, TEXAS

Thomas A. Stidham

Department of Biology, Texas A&M University, 3258 TAMU
College Station, Texas 77843-3258

Fossil vertebrates have been known from the Mioeene of east
Texas for over 100 years (Anonymous 1860; Wilson 1956). Most
of that early work focused on the larger mammalian component of
the fauna. More recently, researchers have examined the smaller
mammals and non-mammalian taxa of the Miocene fauna from east
Texas and Louisiana (Wilson 1956; Albright 1994; 1996; 1998;
Schiebout & Ting 1998; Williams & Schiebout 2003). At present,
the majority of the Miocene vertebrate fossil record in east Texas is
composed of fragmentary teeth and bones. This paper reports what
appears to be the first ichnological record of vertebrate activity in
the Miocene in east Texas.

During road construction along State Highway 19 North of
Huntsville, Walker County, the Texas Department of Transporta¬
tion temporarily exposed rocks belonging to the lower part of the
Fleming Formation (Figure 1) that are stratigraphically near the
contact with the underlying Catahoula Formation (outcropping to
the north of this locality). This temporary outcrop is located 1.29
km (0.8 miles) north of the junction with EM 2821 on State
Highway 19 on the east side of the highway. The outcrop was
examined prior to its covering during road construction. On the
fresh exposure near the top of the outcrop, there were what
appeared to be vertebrate burrows in a weakly developed paleosol.

The outcrop is in the lower part of the Miocene Fleming
Formation near the contact with the underlying Catahoula
Formation. There is a discrepancy between the recognition of
the Fleming contact with the Catahoula in Texas and Louisiana

318

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

Figure 1. Stratigraphy of the outcrop of the Fleming Formation showing the distribution
of burrow structures. Unit 4 is approximately Im thick. Unit descriptions are in the
text.

(Albright 1996), with a portion of the Texas Catahoula Formation
plaeed within the Fleming Formation in Louisiana. Despite this
confliet, the Fleming Formation outerop diseussed here would be
placed within the Fleming Formation in both states. Based on
mammalian faunas published from other Fleming outcrops and
mammalian biostratigraphy (Albright 1998), it appears that this
particular outcrop (in the lower part of the Fleming Formation)
likely is within the Hemingfordian North American Land Mammal
Age in the early Miocene. The absence of mammalian fossils in the
outcrop makes that age determination tentative.

Just under 3.0 meters of section was exposed that had not been
obliterated by a bulldozer (Figure 1). The lowermost unit was
partially covered and only about 40 cm of it was exposed. Unit 2
(approximately 60 cm thick) is a light gray mudstone with nodular

TEXAS J. SCI. 59(4), NOVEMBER, 2007

319

caliche and silicified root traces. Unit 2 contains the potential
vertebrate burrows. Unit 3 (approximately 70 cm thick) has small
linear distributions of caliche nodules. The top of unit 3 appears to
have been the soil surface at the time the burrows were formed.
Unit 4 is a gray mudstone without caliche or root traces and is
approximately one meter thick. Capping these units is unit 5, a thin
10 to 25 cm thick white sandstone with what appears to be small
crustacean burrows near its upper surface.

Unit 2 has many structures that appear to be vertebrate burrows
(Figures 1 and 2). These probable burrows are approximately 1 m
below the old soil surface (top of unit 3). Many of the structures
have large nodular caliche (over 1 cm in diameter) below them.
The structures were recognizable from a distance from the outcrop
because they were a darker gray than the surrounding matrix. The
structures appear to have had a vertical or nearly vertical entrance
and below a subhorizontal chamber. The burrow chamber
structures are approximately 30 cm across horizontally and about
20 cm deep vertically at their deep end. The structures have an
infilling of very finely laminated mudstone. In some instances,
individual internally laminated clay clasts can be seen that appear to
have fallen into the burrow when it was open and are randomly
oriented. One burrow structure has a concentration of silicified root
traces at its top edge as if the roots grew while the burrow was open
to air. The structures roughly resemble a sock. The shape, infilling
with laminated clasts, and presence in a paleosol with caliche
(indicating subaerial exposure) appear to support the identification
of these structures as burrows.

There are a large number of potential burrow makers. The size
and subaerial nature of the burrows makes crustaceans or other
aquatic invertebrates unlikely, and vertebrates seem to be likely
candidates. However, no vertebrate remains were found in the
outcrop. Sediment samples were collected from each unit and the
burrow structures for potential phytolith or palynological analysis.
Nearly any vertebrate including turtles and many rodents or other

320

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

Figure 2. Close-up of a burrow structure infilling from unit 2 showing the difference in
texture and shading of the infilling versus the adjacent matris.

small mammals may have made the burrows. There are a large
number of rodent taxa and other small mammals known from the
Fleming Formation (Albright 1996; Sehiebout & Ting 1998) and
many of these taxa, including heteromyid and geomyoid rodents,
burrow today. It is likely that other burrows or ichnological structures
formed by vertebrates exist in eastern Texas but have been
unrecognized or unobserved due to limited outcrop area and exposure.

Acknowledgments

I thank the Texas Department of Transportation for permission to
examine the outcrop during the construction phase of work along the
highway. Thomas Yancey and John Stidham provided valuable
assistance in the field. Thomas Yancey, Judith Sehiebout, and Pamela
Stephens provided valuable comments on an earlier draft of the
manuscript.

Literature Cited

Albright, L. B. 1994. Lower vertebrates from an Arikareean (earliest Miocene)
fauna near the Toledo Bend Dam, Newton County, Texas. J. Paleo., 68(5): 1 131-
1145.

TEXAS J. SCI. 59(4), NOVEMBER, 2007

321

Albright, L. B. 1996. Insectivores, rodents, and carnivores of the Toledo Bend Local
Fauna: an Arikareean (earliest Miocene) assemblage from the Texas Coastal
Plain. J.Vertebrate Paleo., 26(3):458-473.

Albright, L. B. 1998. The Arikareean Land Mammal Age in Texas and Florida:
southern extension of Great Plains faunas and Gulf Coastal Plain endemism. Pp.
167-183, in Depositional environments, lithostratigraphy, and bio stratigraphy of
the White River and Arikaree Groups (Late Eocene to Early Miocene, North
America). (Terry, D.O., LaGarry, H.E., & Hunt, R.E., eds.). Geological Society
of America Special Paper 325, 216pp.

Anonymous. 1860. Mr. Ashmead in the chair. Proc. Acad. Nat. Sci., Philadelphia,
12(10):416.

Schiebout, J. A. & S. Ting. 1998. Recovery of Miocene terrestrial microvertebrates
from the Fleming Formation in East Texas. Texas J. Sci., 50(3): 199-204.

Williams, M. J. and J. A. Schiebout. 2003. Miocene lower vertebrates from Fort
Polk, LA: a preliminary report. Trans. Gulf Coast Assoc. Geol. Soc., 53:859-865.

Wilson, J. A. 1956. Miocene formations and vertebrate biostratigraphic units, Texas
coastal plain. Bull. Am. Assoc. Petroleum Geol., 40(9):2233-2246.

TAS at: furcula@mail.bio.tamu.edu

"k "k -k it "k

A NOTEWORTHY INFECTION OF
CLINOSTOMUM COMPLANA TUM (DIGENEA: CLINOSTOMIDAE)
IN A CAVE SALAMANDER, EURYCEA LUCIFUGA
(CAUDATA: PEETHODONTIDAE),

FROM NORTHCENTRAL TENNESSEE

Chris T. McAllister, Charles R. Bursey, Matthew L. Niemiller
and Brian T. Miller

Department of Physical & Life Sciences, C hadron State College
Chadron, Nebraska 69337

Department of Biology, Pennsylvania State University-Shenango Campus
Sharon, Pennsylvania 16146

Department of Ecology & Evolutionary Biology, University of Tennessee-Knoxville
Knoxville, Tennessee 37996 and
Department of Biology, Middle Tennessee State University
Murfreesboro, Tennessee 37132

The cave salamander, Eurycea lucifuga inhabits karst areas of
the Central Highlands of Arkansas, Kansas, Missouri, and
Oklahoma east through the Appalachians of Kentucky and

322

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

Tennessee to Virginia and West Virginia (Conant & Collins 1998).
It prefers the twilight zone of eaves and oeeasionally is found near
springs and seepage areas in riparian habitat. This salamander has
been the subject of several parasite surveys, including those of Dyer
& Brandon (1973) in southern Illinois, Dyer & Peck (1975) in
Alabama and Tennessee, Castle et al. (1987) in western Kentucky,
and McAllister et al. (2001) and McAllister & Bursey (2004) in
northcentral Arkansas. Herein is presented new host and distribu-
ional data on a trematode parasite of this salamander from
Tennessee.

On 28 May 2006, an adult male E. lucifuga (SVL = 62 mm) was
collected by hand from the aphotic zone of Flat Rock Cave
(Tennessee Cave Survey Number SM66) in Smith County,
Tennessee (36° 14.06’N, 86° 05.57’W). On closer examination, an
11.1 by 4.3 mm oval-shaped bloody lesion was noted on the right
upper quadrant of the integument of this salamander (Fig. 1). This
salamander was euthanized with a dilute Chloretone® (chlorobu-
tanol) solution and the lesion was excised to reveal an unknown
helminth, 2.5 mm deep within a capsule. The parasite was fixed in
10% buffered formalin, transferred to 70% ethanol, stained with
Semichon’s acetocarmine, and mounted in Canada Balsam. A
voucher specimen of the helminth was deposited in the United
States National Parasite Collection (USNPC), Beltsville, Maryland,
USA, as USNPC 98667. A host voucher specimen was deposited
in the Middle Tennessee State University Collection, Murfreesboro,
Tennessee, USA, as MTSU 358C.

On further examination, the helminth was identified as a
metacercaria of Clinostomum complanatum (Rudolphi 1814). The
identification is based on the morphology of cuticular spines as
follows: in Clinostomum attenuatum, cuticular spines measure 13-
16 jam in length by 5-9 jam in thickness; in Clinostomum
complanatum, cuticular spines measure 7-11 jam in length by 1.5-2
jam in thickness (Cort 1913). The cuticular spines of this specimen
fall within the latter range. Interestingly, a total of only 1 of 549
(0.18%) E. lucifuga (391 adults, 25 juveniles, and 133 larvae) from

TEXAS J. SCI. 59(4), NOVEMBER, 2007

323

Fig. 1. Adult Eurycea lucifuga showing encysted metacercaria of Clinostomum

complanatum (arrow).

the current site and 67 other localities in Tennessee harbored this
worm. This represents a new host and distributional record for C.
complanatum.

Metacercariae of C. complanatum have been previously reported
in other North American amphibians, including three species of
ranid frogs (Manter 1938; Walton 1949; Fried & Foley 1970) and
five species of salamanders (summarized in Table 1). In addition,
metacercaria of Clinostomum attenuatum has been reported from
five of eight (62.5%) barred tiger salamanders, Ambystoma
tigrinum mavortium in Texas (Miller et al. 2004), and a
Clinostomum sp. was reported from 12 of 163 (7.4%) eastern newts,
Notophthalmus viridescens from Michigan (Muzzall 1991).

Dermal cysts of Clinostomum sp. metacercariae have been
reported on the head and body of a spotted salamander, Ambystoma
maculatum from Maryland (Fowler 1947). Indeed, metacercariae
of Clinostomum spp. are typically found in large, raised cysts
randomly distributed on or within the host’s body; the cysts develop
as surface structures and do not infiltrate adjacent tissues (i.e.,
significant pathological changes are rarely associated with the cysts,
although occasionally increased numbers of melanomacrophages
surround a cyst) (Miller et al., 2004). A similar pathology was
observed in the host noted herein (see Fig. 1).

324 THE TEXAS JOURNAL OE SCIENCE-VOL. 59, NO. 4, 2007

Table 1. Metacercaria of Clinostomum complanatum in North American salamanders.

Eamily/Host Species

Locality

Prevalence*

Reference

Amphiumidae

Amphiuma tridactylum

Louisiana

6/85 (7.1%)

Bennett &

Plethodontidae

Eurycea lucifuga

Tennessee

1/549 (0.2%)

Humes (1938)

This study

Eurycea neotenes

Texas

1/86(1.2%)

McAllister (1990)

Salamandridae

Notophthalmus viridescens

Massachusetts

1/1 (100%)

McAllister (1990)

Sirenidae

Siren sp.^

Elorida

1/1 (100%)

Manter(1938)

' Number infected/number examined (percent).

^ Host is thought to be a greater siren, S. lacertina (see Bennett & Humes 1938; Eowler
1947).

In the life cycle of Clinostomum spp., embryonated eggs pass in
the feces of the definitive host and miracidia hatch and penetrate
tissues of a snail (first intermediate host). Sporocysts and two
generations of rediae develop producing cercariae that eventually
encyst in fishes or amphibians (second intermediate hosts). The
adult worm develops in the mouth or esophagus of various
piscivorous birds (herons, gulls, and bitterns) that serve as
definitive hosts (Hopkins 1933; Hunter & Hunter 1933; Schell
1985; Aohagi et al., 1992; Dias et al. 2003). It is unknown what
specific host life cycle is followed in this cave environment since
other potential second intermediate hosts are found within,
including southern two-lined salamanders, Eurycea cirrigera,
northern green frogs, Lithobates {Rand) clamitans melanota,
American bullfrogs, Lithobates (Rana) catesbeianus, and the
southern cavefish, Typhlichthys subterraneus . Indeed, no definitive
bird hosts have ever been observed at the cave entrance (situated in
a wooded ravine) although domestic cattle seeking a cooler
environment frequent the entrance during summer.

Acknowledgments

We thank the Tennessee Wildlife Resources Agency for
Scientific Collecting Permit No. 1450 issued to B.T.M. and the
private landowner for allowing access to Flat Rock Cave.

TEXAS J. SCI. 59(4), NOVEMBER, 2007

325

Literature Cited

Aohagi, Y., T. Shibahara, N. Machida, Y. Yamaga, K. Kagota & T. Hayashi. 1992.
Natural infections of Clinostomum complanatum (Trematoda: Clinostomatidae)
in wild herons and egrets, Tottori Prefecture, Japan. J. Wildl. Dis., 28:470-471.

Bennett, H. J. & A. G. Humes. 1938. Helminth parasites of Amphiuma tridactylum
and Siren lacertina from Louisiana. Proc. Louisiana Acad. Sci., 4:243-245.

Castle, M. D., D. A. Strohlein & B. M. Christensen. 1987. Helminth parasites of the
cave salamander, Eurycea lucifuga, from western Kentucky. Proc. Helminthol.
Soc. Washington, 54:269-270.

Conant, R. & J. T. Collins. 1998. A field guide to reptiles and amphibians of eastern
and central North America, 3*^^ Edition, expanded. Houghton-Mifflin, Boston,
Massachusetts, 616 pp.

Cort, W. W. 1913. Notes on the trematode genus Clinostomum. Trans. American
Microsc. Soc., 32:169-182.

Dias, M. L., J. C. Eiras, M. H. Machado, G. T. Souza & G. C. Pavanelli. 2003. The
life cycle of Clinostomum complanatum Rudolphi, 1814 (Digenea,
Clinostomidae) on the floodplain of the High Parana River, Brazil. Parasitol.
Res., 89:506-508.

Dyer, W. G. & R. A. Brandon. 1973. Helminths in three sympatric species of cave¬
dwelling salamanders in southern Illinois. Trans. Illinois Acad. Sci., 66:23-29.

Dyer, W. G. & S. B. Peck. 1975. Gastrointestinal parasites of the cave salamander,
Eurycea lucifuga Rafmesque, from the southeastern United States. Canadian J.
Zook, 53:52-54.

Fowler, J. A. 1947. A new host for Clinostomum metacercariae. J. Parasitol.,
33:444.

Fried, B. & D. A. Foley. 1970. Development of Clinostomum marginatum
(Trematoda) from frogs in the chick and on the chorioallantois. J. Parasitol.,
56:332-335.

Hopkins, S. H. 1933. Note of the life history of Clinostomum marginatum
(Trematoda). Trans. American Microsc. Soc., 52:147-149.

Hunter, W.S. & G. W. Hunter, III. 1933. The miracidium of Clinostomum
marginatum (Rud.). J. Parasitol., 20:132.

Manter, H. W. 1938. A collection of trematodes from Florida Amphibia. Trans.
American Microsc. Soc., 57:26-37.

McAllister, C. T. 1990. Metacercaria of Clinostomum complanatum (Rudolphi,
1814) (Trematoda: Digenea) in a Texas salamander, Eurycea neotenes
(Amphibia: Caudata), with comments on C. marginatum (Rudolphi, 1819). J.
Helm. Soc. Washington, 57:69-71.

McAllister, C. T. & C. R. Bursey. 2004. Endoparasites of the dark-sided
salamander, Eurycea longicauda melanopleura, and the cave salamander,
Eurycea lucifuga (Caudata: Plethodontidae), from two caves in Arkansas, U.S.A.
Comp. Parasitol., 71:61-66.

McAllister, C. T., S. E. Trauth & L. W. Hinck. 1991. Sphryanura euryceae
(Monogenea) on Eurycea spp. (Amphibia: Caudata), from northcentral Arkansas.
J. Helminthol. Soc. Washington, 58:137-140.

326

THE TEXAS JOURNAE OF SCIENCE-VOL. 59, NO. 4, 2007

Miller, D. L., C. R. Bursey, M. J. Gray & L. M. Smith. 2004. Metacercariae of
Clinostomum attenuatum in Ambystoma tigrinum mavortium, Bufo cognatus and
Spea multiplicata from west Texas. J. HelminthoL, 78:373-376.

Muzzall, P. M. 1991. Helminth infracommunities of the newt, Notophthalmus
viridescem, from Turkey Marsh, Michigan. J. ParasitoL, 77:87-91.

Schell, S. C. 1985. Handbook of trematodes of North America north of
Mexico.University Press of Idaho, Moscow, Idaho, 263 pp.

Walton, A. C. 1949. Parasites of the Ranidae (Amphibia). XIV. Tran. Illinois St.
Acad. Sci., 42:161-164.

CTM at: cmcallister@csc.edu

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

327

INDEX TO VOLUME 59 (2007)
THE TEXAS JOURNAL OF SCIENCE

Jeffery A Masters

Department of Biology, Angelo State University
San Angelo, Texas 76909

This index has separate subject and author sections. Words, phrases,
locations, proper names and the scientific names of organisms are followed by the
initial page number of the articles in which they appeared. The author index
includes the names of all authors followed by the initial page number of their
respective article(s).

SUBJECT INDEX

Acer negundo 243
Atlantic ghost crab 5 1
American beaver 243

Balcones Fault Zone 163
Big Hill Bayou 33
black willow 243

Blanco River 163, 179, 201, 209, 233

Bolivar Peninsula 5 1

box elder 243

Bryopsidales 305

buckling capacity 97

bullfrogs 33

bur oak 243

burrows 11,23,51,317

Cagle’s map turtle 201
Carabidae 39
Carnivora 137
cardiac tissue 277

Carya illinoensis 243
Castor canadensis 243
Castoridae 243
Caudata 321
cave salamander 321
cedar elm 243
Celtis sp. 243
Central America 3 1 1
chick heart development 277
chigger mites 23
Chihuahuan Desert 39
Chinaberry 243
Chiroptera 137
Chlorophyta 103, 305
Chondrophycus papillosa 103
Chrysemys dorsal is 155
Clinostomidae 321
Clinostomum complanatum 321
Coleoptera 39
Colubridae 3 1 1
Croton punctatus 1 1 3
Crustacea 291
Cyprinella venusta 179
Cyprinids 179

328

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

Dalquest Research Site 39
dedolomitization 1 63
deer management 26 1
desert lizards 23
diagenesis 163
Digenea 321

digital image analysis 103
Dipodomys elator 1 1
dune habitat 23
dune vegetation 1 1 3

eastern cottonwood 243
Edwards Aquifer 163
Emydidae 155
Eqiius asinus 3
Eurycea hicifuga 321
Eiitrombicula alfreddugesi 23

feral burro 3
foraging preferences 243
Fleming Formation 317
Florideophyceae 103
fibroglycan 277
fish assemblages 179

Graptemys caglei 20 1
Great-tailed grackle 151
growth factors 277
Guadalupe River 73,201

hackberry 243
heart development 277
helminth parasites 33
Hemidactylus turcicus 127
Heterotheca subaxillaris 1 1 3
heparin sulfate proteoglycans 277

hollow square pipe columns 97
honey mesquite 1 1

Ipomoea pes-caprae 1 1 3
immunohistochemical analysis 277

karst stream 1 63

Lagomorpha 137
Laguna Madre, lower 305
Laurencia papillosa 103
lotebush 1 1

macroalgae 103, 305
Mad Island Marsh Preserve 61
Marsupial 137
Mediterranean gecko 127
Melia azedarach 243
mesquite-juniper association 137
Mexican buckeye 243
Mexico 103

northeastern 137
northern 23
southern 3 1 1
Cities

Aramberri 137

Viesca 23
States

Nuevo Leon 137
Miocene 317
Morus alba 243
muricatacin 301

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

329

Nature Conservancy, the 233
Neoechinorhynchus 33
Neotoma albigula 1 3 7
Ninia sebae 3 1 1
Northern blot analysis 277
Nototropis amabilis 179

Ocypode quadrata 5 1
Odocoileus virginianus 26 1
Ostracoda 29 1

pecan 243

Penicillm capitatus 305
Peromyscus leucopus 1 3 7
P eromy sens pector alls 137
Phaeophyceae 103
Pimephales vigilax 179
Plethodontidae 321
playa lakes 291
Poecilia latipinna 73
Populus deltoides 243
Prosopis glandulosa 1 1
PuntaYuYum 103

Quercus macrocarpa 243
Quiscalus mexicanus 1 5 1
Quintana Roo 103

Rana catesbeiana 33
redback coffee snake 3 1 1
Rodentia 137,243

sailfin molly 73
Salix nigra 243
San Antonio River 201
San Juan y Puentes 137
San Marcos River 163,201
Scarabaeoidea 39
seed shrimp 29 1
selective foraging 243
Serpentes 3 1 1
shoreline change 61
southern painted turtle 155
South Padre Island 1 13
Spring Lake 201
syndican-2 277

tartaric acid 301
Tenebrionoidea 39
Tennessee 321
Testudines 155
Texas 51,113,155

central 73,163,209,233,243
Edwards Plateau 163,233
Hill Country 163
Llano Estacado Region 291
northwest 291
south 305,261
southeast 33
Southern High Plains 291
Trans-Pecos 3, 39
west 3, 39
Cities

Fisher 201
Hunstville 317
San Marcos 201

330

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

Counties

Gonzalez County 201
Guadalupe County 201
Hays County 201
Jefferson County 33
Matagorda County 6 1
Walker County 317
Wichita County 1 1
Texas kangaroo rats 1 1
Trematoda 371

triangle preserving functions 83
Trinidad y Llano Salas 137
Trinity Aquifer 163
Triticum aestivum 119
Turbinaria tricostata 103
Twin Falls Volcanic Field 3

Ulmus crassifoilia 243
Uma exsul 23
Ungnadia speciosa 243
ungulates 261
Uta stejnegeri 23

water distribution 26 1
white mulberry 243
White-tailed deer 261

Yucatan Peninsula 103
Z

Zizyphus obtusifolia 1 1

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

331

AUTHORS

Athey, J. 73

Baccus, J. T. 243
Barrows, C. W. 23
Bean, P. T. 179
Bonner, T.H. 179
Boza, G. H., Jr. 305
Broussard, G. H. 3, 39
Bursey, C. R. 321

Castaneda, G. 23
Cave, M. S. 163
Contreras- Arquieta, S. 137
Contreras-Balderas, A. 137
Cook, W. B. 3
Curran, J. C. 209

DeYoe,H. R. 305

Faruqi, M.A. 97
Fikes, R. L. 103
Forstner, M. R. J. 155
Fuller, J.P. 155

Gammon, D. E. 151
Garda-De la Pena, C. 23
Goetze, J. R. 11
Goldberg, S. R. 311
Gomez, G. W. 33
Gordon, A. 277
Groeger, A. W. 163

Halstead, L. E. 233
Hellickson, M. W. 261
Hewitt, D.G. 261
Hockaday, D. L. 305
Home,F. R. 291
Homer, N. V. 3

Jamssi, A. S. 73
Jester, S. 233
Judd,F.W. 113

Kainer, M. A. 243
Kowalski, J. L. 305

Langford, J. K. 277

Lehman, R. L. 1 03
Littrell, B. M. 179
Lonard,R. 1. 113

Maccarone, A. D. 51
Mangham, W. 6 1
Mathews, P. L. 51
Mazariegos, R. A. 113
McAllister, C. T. 155,321
Middleton, S. M. 39
Miller, B.T. 321

Nelson, A. D. 11
Niemiller, M. L. 321

Rose, F. L. 201

Sears, N.L. 301
Shah, P. K. 97
Shipley, M.M. 39
Simpson, T. R. 201
Small, M.F. 243
Smith, L.C. 103
Southard, J. M. 301
Spellman, J. W. 83
Stangl, Jr. F. B. 3
Stasey, W. C. 11
Stevens, F. B. 73
Stidham, T. A. 317
Strenth,N. E. 291
Sudman, P. D. 11
Summy, K. R. 113

Torrejon, R. M. 83
Torres-Ayala, J.M. 137

Valero-Padilla, J. 137
Vogtsberger, R. C. 39

Wampler, E.S. 151
Webb, S. L. 261
Wiebusch, P. L. 127
Williams, H. 61

Yoder, H.R. 33

ZaidanIII,F. 127

332

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

REVIEWERS

The Editorial staff wishes to acknowledge the following individuals
for serving as reviewers for those manuscripts considered for publica¬
tion in Volume 59. Without your assistance it would not be possible to
maintain the quality of research results published in this volume of the
Texas Journal of Science.

Nathan Allan
Keith Arnold
John Baccus
Nathan Bell
Wayne Bennett
Guy Bernard
Troy Best
David Bolender
Brent Burt
Alejandro Castillo
Rodney Cate
Brian Chapman
Randy Combs
Melinda Coogan
Tami Cook
William Cook
Donald Davies
William Davis
Timothy de la Pena
Hudson DeYoe
James Dixon
Robert Dowler
Norman Dronen
Kenneth Dunton
Bart Durham
Robert Edwards
Regina Edziye
Carl Ernst
William Font
Gary Garrett
Kenneth Gee
James Gibeaut
Jim Goetze

Alan Groeger
Ranette Halverson
Christopher Hansen
Don Harper
Toby Hibbitts
Christine Hice
David Hieks
Ronald Hill
Jeff Hood
John Janovy, Jr.

Richard Kazmaier
Falvius Killebrew
Marie King
Robert Kremer
Travis LaDuc
Tom LaPoint
John Lewis
Steve Lingafelter
Lance Littlejohn
William Lutterschmidt
Edie Marsh-Mathews
James Masuoka
Geoffrey Mathews
J. Kelly McCoy
John McEachran
W. Ted Mehavier
Dana Mills
Patrick Mitchell
Michael Morrison
Patrick Muzzall
Allan Nelson
Gerald North
Darlene Occena-Gutierrez

Vatsala Padapalli
Dennis Parmley
C.O. Patterson
Larry Peel
Mike Quist

Magaly Rincon-Zachary
Eric Rominger
Francis Rose
William Ross
Jon Scales
Judith Schiebout
Joel Schmitter
John Schwarz
Michael Shipley
Richard Simpson
John Sivinski
Jaime Slye
Michael Small
Gerald Smith
Frederick Stangl
Wayne Starnes
Pamela Stephens
Fred Stevens
Ned Strenth
A1 Sutko

William Van Auken
Roy Vogtsberger
George Ward
Ken Wilkins
Kim Withers
Yixin Zhang
Gregory Zolnerowich

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

IN RECOGNITION OF THEIR ADDITIONAE SUPPORT OF
THE TEXAS ACADEMY OF SCIENCE DURING 2007

PATRON MEMBERS

Ali R. Amir-Moez
Stephen R. Goldberg
Deborah D. Hettinger
Don W. Killebrew
David S. Marsh
John Sieben
Ned E. Strenth

SUSTAINING MEMBERS

James Collins
Donald E. Harper, Jr.
Michael Looney
Fred Stevens
William F. Thomann
Milton W. Weller
Kristin Miller White

SUPPORTING MEMBERS

Samuel C. Caran
David Rolf Cecil
George D. McClung
Jim Neal
William J. Quinn
Jack B. Robinson

334

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

THE TEXAS ACADEMY OF SCIENCE

www.texasacademyofscience.org

Membership Information and Application

MEMBERSHIP.-Any person or member of any group engaged in scientific
work or interested in the promotion of science is eligible for membership in The
Texas Academy of Science.

(Please print or type)

Name _

Last First Middle

Mailing Address _

City _ State _ Zip _

Ph _ FAX _ E-mail: _

Regular Member $30.00 _ Emeritus $10.00 _

Supporting Member $60.00 _ Corporate Member $150.00 _

Sustaining Member $100.00 _ Affiliate $5.00 _

Patron Member $ 1 50.00 _ (list name of organization)

Student-Undergraduate $15.00 _ _

Student-Graduate $15.00 _ Contribution _

Joint $35.00 _ AMOUNT REMITTED _

SECTIONAL INTEREST AREAS:

Anthropology

Botany

Cell and Molecular Biology
Chemistry and Biochemistry
Computer Science
Conservation Ecology
Environmental Science

Freshwater & Marine Sciences

Geosciences

Mathematics

Physics

Science Education

Systematics & Evolutionary Biology

Terrestrial Ecology & Management

Please indicate your Sectional interest(s) below:

1. _ 2. _ 3.

Send Application Form and Check or Money Order to:

Dr. Fred Stevens, Executive Secretary
The Texas Academy of Science
CMB 6252
Schreiner University
Kerrville, Texas 78028-5697

Please photocopy this Application Form

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

335

United Stales Postal Sen^e

Statement of Ownership, Management, and Circulation

1. Publication Title

The Texas Journal of Science

2. Publication Number

3. Filing Dale

1 October 2006

0040_4A 03

4. Issue Frequency

Quarterly

5. Number of Issues Published Annually

$. Annual Subscription Price

$30 Membership
$50 Subscription

7. Complete Mailing Address of Known Office of Publication (Not printer) (Street, city, county, state, andZiP-rA)
Biology Department, Angelo State Ufiiversity

2601 West Avenue N, Tom Green County, San Angelo, TX 76909

Contact Person

N.E. Strenth

Telephone

325.942.2189 ex. 247

8. Complete Mailing Address of Headquarters or General Business C5ffice of Publisher (Not printer)

Dr. Ned E. Strenth, Biolo^r Department

Angelo State University

San Angelo, IX 76909-5069 _ _ _

9. Full Names and Complete Mailing Addresses of Publisher, Editor, and Managing Editor (Do not leave blank) _ _ _

PubPisher (Name and complete mailing address)

Dr. Ned E. Strenth, Biology Department
Angelo State University

San Angelo. TX 76909-5069 _ _ _

Editor (Name and complete mailing address)

Dr. Ned E. Strenth, Biology Department
Angelo State Iftiiversity

San Angelo, TX 76909-5069 _ _ _ _ _

Managing Editor (Name and compiete mailing address)

Dr. Ned E. Strenth, Biology Department
Angelo State University

San Angelo, TX 76909-5069 _ _ _ _ _

10. Owner (Do not leave blank. If the publication is owned by a corporation, give the name and address of the corporation immediately followed by the
names and addresses of all stockholders owning or holding 1 percent or more of the totai amount of stock Hnot owned by a corporation, give the
names and addresses of the individual owners, tf owned by a partnership or other unincorporated firm, give its name and address as well as those of
each individual owner. If the publication is pubiished by a nonprofit organization, give its name and address.)

Full Name

Complete Mailing Address

Texas Academy of Science

Angelo State University Department of Biolog

2601 West Avenue N

San Angelo, TX 76909-5069

11. Known Bondholders, Mortgagees, and Other Security Holders Owning or
Holding 1 Percent or More of Total Amount of Bonds, Mortgages, or

Other Secunties. If none, check box - ► Q None

Full Name

Complete Mailing Address

1 2. Tax Status (For completion by nonprofit organizations authorized to maii at nonprofit rates) (Check one)

The purpose, function, and nonprofit status of this organization and the exempt status for federal income tax purposes:
CXHas Not Changed During Preceding 12 Months

□ Has Changed During Preceding 12 Months (Pubiisher must submit explanation of change with this statement)

PS Form 3526, October 1 999

(See Instructions on Reverse)

336

THE TEXAS JOURNAL OF SCIENCE-VOL. 59, NO. 4, 2007

13. Publication Title

The Texas Journal of Science

14. Issue Date for Circulation Data Below

November 2006

^ Extent and Nature of Circulation

Average Na Copies Each Issue
During Preceding 12 Months

No. Copies of Single Issue
Published Nearest to Hling Date

a. Total Number of Copies (Net press nin)

1100

b. Paid and/or
Requested
Circulation

(1)

Pakl/Requesled Outside-County MaB Subscriptions Stated on
Form 3541 . (Include advediser's proof and exchange copies)

808

965

(2)

Paid In-County Subscriptions Staled on Form 3541
(Induda advertiser’s proof and exchange copies)

(3)

Sales Through Dealers and Carriers, Street Vendors,
Counter Sales, and Other Non-USPS Paid DistnTjution

157

(4)

Other Classes Mailed Through the USPS

Total Paid and/or Requested Circulation w

(Sum of ISb. (1). (2).(3),and(4)] ~

980

''Free
Distribution
by Mail
(Samples,
compliment
ary, and
other tree)

Outside-County as Staled on Form 3541

(2)

In-Counly as Stated on Form 3541

(3)

Other Classes Mailed Through the USPS

Free Distribution Outside the Mail
(Carriers or other means)

Total Free Distn^bution (Sum of 15d. and 15e.) ^

g. w

TotaJ Distribution (Sum of 15c. and 15f) ^

980

Copies not Distributed

120

Total (Sum of 15g. andh.) ^

1100

i- Percent Paid and/or Requested Circulation
(15c. divided by 15g. times 100)

89%

16. Publication ot Statement of Ownership

Publication required. Will be printed in the

17. Signature and Title of Editor, Pul

Vol. 58, #4

. issue of this publication.

□ Publication not required.

Title of Editor, PuMeher;

Business Manager, or Owner

29 Sept. 2006

I certify that all infomiation furnished on this form Is true and complete. I understand that anyone who furnishes false or misleading infomnation on this form
or who omits materia) or information requested on the form may be subject to criminal sancfions pnduding fines and Imprisonme nt) and/or dvil sanctions
(including civil penalt'es). '

Instructions to Publishers

1 . Complete and file one copy of this form with your postmaster annually on or before October 1 . Keep a copy of the completed form
for your records.

2. In cases where the stockholder or security holder is a trustee, include in items 1 0 and 1 1 the name of the person or corporation for
whom the trustee is acting. Also include the names and addresses of indMduals who are stockholders who own or hold 1 percent
or more of the total amount of bonds, mortgages, or other securities of the publishing corporation. In item 11, if none, check the
box. Use blank sheets if more space is required.

3. Be sure to furnish all circulation information called for in item 15. Free circufation must be shown in items 15d, e, and f.

4. item 15h., Copies not Distributed, must include (1) newsstand copies originally stated on Form 3541 , and relumed to the publisher,
(2) estimated returns from news agents, and (3), copies for office use, leftovers, spoOed, and all other copies not distribute.

5. If the publication had Periodicals authorization as a general or requester publication, this Statement of Ownership, Management,
and Circulation must be published; it must be printed in any issue in October or, if the publication is not published during October,
the first issue printed after October.

6. In item 1 6, indicate the date of the issue in which this Statement of Ownership vrill be published.

7. Hem 1 7 must be signed.

Failure to file or publish a statement of ownership may lead to suspension ot Periodicals authorization.

PS Form 3526, October 1999 (Reverse)

THE TEXAS ACADEMY OF SCIENCE, 2007-2008

OFFICERS

President'.

President Elect
Vice-President.

Immediate Past President.
Executive Secretary'.
Corresponding Secretary:
Managing Editor:

Manuscript Editor:

Treasurer:

AAAS Council Representative:
International Coordinator:

Hudson R. DeYoe, University of Texas-Pan American

Raymond C. Mathews, Jr., Texas Water Dev. Board

William J. Quinn, St. Edward’s University

David S. Marsh, Angelo State University

Fred Stevens, Schreiner University

Cindy Contreras, Texas Parks and Wildlife Department

Ned E. Strenth, Angelo State University

Frederick B. Stangl, Jr., Midwestern State University

John A. Ward, Brooke Army Medical Center

James W. Westgate, Lamar University

Armando J. Contreras, Universidad Autonoma de N.L.

DIRECTORS

2005 Jerry L. Cook, Sam Houston State University
Flo Oxley, Lady Bird Johnson Wildflower Center

2006 Herbert D. Grover, Hardin- Simmons University
Gary P. Garrett, Texas Parks and Wildlife Department

2007 Renard L. Thomas, Texas Southern University
Bob Murphy, Texas Parks and Wildlife Department

SECTIONAL CHAIRPERSONS

Anthropology: Roy B. Brown, Instituto Nacional de Antropologia y Historia

Botany: Alan W. Elevens, Texas Lutheran University

Cell and Molecular Biology: Jon B. Scales, Midwestern State University

Chemistry and Biochemistry: Benny E. Amey, Jr., Sam Houston State University

Computer Science: Laura J. Baker, St. Edward’s University

Conservation Ecology: Cathy Early, University of Mary Hardin Baylor

Environmental Science: Kenneth R. Summy, University of Texas-Pan American

Ereshwater and Marine Sciences: Brian W. Brooks, Baylor University

Geosciences: Carol Thompson, Tarleton State University

Mathematics: Sandra Luna McCune, Stephen F. Austin State University

Physics: David Bixler, Angelo State University

Science Education: R. Russell Wilke, Angelo State University

Systematics and Evolutionary Biology: Allan W. Hook, St. Edward’s University

Terrestrial Ecology and Management: Christopher M. Ritzi, Sul Ross State University

COUNSELORS

Collegiate Academy: William J. Quinn, St. Edward's University
Junior Academy '.W'mcQ Schielack, Texas A&M University

SMITHSONIAN INSTITUTION LIBRARIF*;

3 9088 01431 9081

THE TEXAS JOURNAL OF SCIENCE PERIODICALS

Texas Academy of Science

CMB 6252

Schreiner University

Kerrville, Texas 78028-5697

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Texts

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "The Texas journal of science"

See other formats